White, C. (In press). Cross-cultural Similarities in Reasoning about Personal Continuity in...

DRAFT VERSION. FORTHCOMING IN RELIGION, BRAIN & BEHAVIOR

Cross-Cultural Similarities in Reasoning About Personal Continuity in

Reincarnation: Evidence from South India

Claire White

Department of Religious Studies, California State University, Northridge, USA

email: claire.white@csun.edu

Abstract

Ethnographic reports suggest the cross-cultural recurrence of practices designed to determine who

has returned to the human world through the process of reincarnation. This paper attempts to

explain these trends through the cross-cultural replication of experiments with participants from

mixed-religious affiliations in the U.S. to Jains in South India. In a series of imaginative

perspective-taking tasks, Jain adults reasoned about the likelihood of different people to be the

reincarnation of a deceased person based on similar features between the two. Participants

endorsed a concept of reincarnation as entailing a bodily change from death to rebirth.

Nevertheless, in the tasks, they reasoned as though physical marks and episodic memories were

superior cues to identifying reincarnated people as compared to other similarities, including other

physical and mnemonic features. These decisions were heightened when the distinctiveness of the

features were increased and thus the probability of another living person sharing that feature

decreased. Together, these studies replicate and extend findings with U.S. adults to suggest that

such reasoning is not simply the product of cultural input and suggest a role for mundane processes

of person recognition in the recurrence of reincarnation concepts and associated practices cross-

culturally.

Keywords

Personal identity, reincarnation, Jainism, cross-cultural, social cognition

Cross-cultural Similarities

1. Introduction

In many societies, biological birth does not mark a person’s entrance into the world, but rather,

their return to it from a previous life in another bodily form. Cross-cultural surveys that adopt this

conceptualization of reincarnation have documented the belief in around 30% of cultures (e.g., see

Matlock, 1993; Obeyesekere, 2002; Somersan, 1981; Swanson, 1960) including: Melanesia, South

Asia, North America, Australia and Europe; and especially: West Africa, India, and Northwestern

North America (Matlock, 1993; Obeyesekere, 1994). In such contexts, people often believe that a

particular person, such as a beloved kinsman, come back to live among them in the human world

through rebirth1. One pertinent question is how do people determine who has returned? This question

is important for many reasons. For one, in reincarnationist traditions, decisions about who has returned

to the human world have huge social consequences: they determine the distribution of names, titles,

property inheritance, wealth distribution, child-rearing practices and, in the case of Tibet, even

religious governance (see Gottlieb, 2004; Kaplan, 2000; White, Sousa & Berniunas, 2014).

Ethnographic reports documenting practices designed to determine who has returned to the

human world within and across cultures (e.g., for North America, see Matlock & Mills, 1994), suggest

that these practices are remarkably similar: people often seek out and interpret two types of similarities

between the person before death, and after rebirth, as evidence that they are one and the same person:

namely, physical marks and memories (e.g., Bastian, 2002; Bell, 1931; Creider, 1986; Gottlieb, 2004;

Gupta, 2002; Haraldsson & Samararatne, 1999; Mills & Slobodin, 1994; Obeyesekere, 2002; Oluwole,

1992; Orubu, 2001; for an overview of these reports, see White, 2014). Given ideas about

reincarnation in the surrounding cultural milieu, the popularity of these two similarities in particular is

also surprising. First, although there is cultural variance on what, exactly, remains constant in the

process of reincarnation, there is general cultural convergence that the physical body changes. In other

words, the person before death has a different body than the person after rebirth. Yet in many parts of

the world (especially cultures in India and Africa), newborn children are routinely examined for

1 This research concerns human-to-human reincarnation only, as opposed to the transformation of humans into other

ontological categories (e.g., human-to-animal). Here, the term person is a synonym for human being and the two terms are used interchangeably throughout. The question of what constitutes a person, and what qualifies for “personhood” has a long and controversial history across disciplines (e.g., see Cohen & Barrett, 2011; Macdonald, 2005) and such considerations are beyond the scope of this paper.

Cross-Cultural Similarities

physical marks that correspond to those on the deceased. These are often distinctive physical marks

(i.e., marks that are uncommon or unique), such as: skin abnormalities (e.g., discoloration, lesions,

spots, moles) thought to correspond to birthmarks on the deceased (i.e., marks present from the first

three months of birth) or wounds inflicted to the person at the time of death2. In other contexts

(especially cultures in West Africa and Northwestern North America), the corpses of children who die

in infancy are physically marked to make identification in the next life easier (e.g., Frazer, 1918;

Goody, 1962; Parkin, 1988).

Second, in other contexts, children are often tested for the recognition of places or people the

deceased knew, or items belonging to them, which assumes the continuity of episodic autobiographical

memory (i.e., memory of events that happened to the self, Tulving, 1972; hereafter “episodic

memory”). These ideas are predominant even in reincarnationist traditions where they contradict

corresponding dogma and discourse on the process of reincarnation. In many Buddhist traditions, for

example, the concept of anattā maintains that there is no permanent, unchanging self, yet people seek

out similarities in memory between the deceased, when alive, and living to establish the identity of

reincarnated Lamas and in Tibetan Buddhism, the selection of Dalai Lama partly depends upon it (see

Haraldsson & Samararatne, 1999; Nicholson, 2014; Tenzin, 1990; White, Sousa & Berniunas, 2014).

Together, these observations raise a number of questions. First, why do people, in very different

contexts, seek out, and interpret, similarities between the living and deceased as evidence that the two

people (before death and after rebirth) are numerically identical? In other words, despite the dramatic

physical changes associated with reincarnation, people assume that the individual before death and

after rebirth constitute one, rather than two, persons3. Second, why do people seek out evidence for the

2 Note that physical features that are taken as evidence of reincarnation are often marks present at birth. This may indicate that assumptions that temporal distance between people at two points in the rebirth process affect judgments about continued identity. Namely, that marks present from birth and at death are at the starting and end points of life and are thus stronger indicators of continued identity because they are temporally as close to the reincarnated agent as possible (see Blok, Newman & Rips, 2005, for a discussion of the principle of spatio-temporal continuity and identity). A fuller discussion of this is beyond the scope of this paper. 3 To say that a person is numerically identical is to say that they are one and same thing: one person rather than two. For example, if you take count of how many of your students are in a lecture, and how many return after a short break, you are qualifying and counting each student that returns as numerically identical to a student who left the room previously. By contrast, to say that people are qualitatively identical is to say that they are exactly similar, though they may not necessarily be numerically identical. For instance, if identical twins return to your lecture after the break, but only one was present at the morning session, you have two people who are qualitatively identical to the person who left but only one is numerically identical. The questions in this research concern how people can change through reincarnation (and are no longer qualitatively identical), yet are attributed as being numerical identical to the person before death.

continuity of episodic memory and distinctive physical marks in particular, especially given the

inherent contradictions in corresponding theological discourse on reincarnation? In what follows, we

propose some intuitive cognitive biases involved in decisions about personal continuity in

reincarnation.

1.1 A cognitive approach to reincarnation

This research is the first to consider reincarnation from a psychological perspective, often

referred to as the cognitive science of religion, which assumes that religious ideas are part and

parcel of ordinary cognition. It takes as a starting point the assumption that to explain the spread of

ideas about reincarnation – in addition to socio-historical factors – one needs to consider the role

of human cognition in constraining and facilitating their transmission (e.g., Barrett, 2000; Boyer,

2001, Sperber, 1996). On this account, reincarnated agents – like other supernatural agents; such as

gods, ghosts, and people possessed by spirits – are facilitated and constrained by the intuitive

representation of human agents generally (see Barrett, 1999; 1998; Bering, 2002; Boyer, 2003;

Cohen, 2007; Cohen & Barrett, 2008; Guthrie, 1993). One possibility, therefore, is that these

cross-culturally recurrent ideas – about how to identify reincarnated people – are readily generated,

remembered and communicated in part, because they meet cognitively optimal assumptions about

how to identify individual people from one time to the next: in short, they provide convenient and

reliable evidence of personal continuity4. In the context of rebirth, however, where people also

represent individuals as undergoing dramatic physical change, decisions about reliability may also

be informed by assumptions about what people think constitutes personal identity – and thus what

is likely to continue5.

Distinctive physical marks indicate bodily continuity

In everyday contexts, we ascribe continued identity automatically to people everyday based on

their physical appearances. I assume that the person I greet at the end of the day is the same

person I kissed goodbye at the start of the day because they look similar at both times. However, 4 Here, decisions about personal continuity concern how we attribute numerical identity to others (i.e., that they constitute one, rather than two persons) despite dramatic qualitative changes (e.g., occupying a new body). 5 Personal identity here refers to the property, set of properties or quality that the folk think makes a person who they are and enables them to continue over time (e.g., see Butler, 1819; Kaufman, 2004; Locke, 1690/1975; Nagel, 1986; Perry, 1978; Reid, 1785/1969; Shoemaker & Swinburne, 1984; Williams, 1973).

decisions about how to recognize a person over time (personal continuity) are not the same as

decisions about what we think constitutes an individual’s identity6. For example, I assume that the

person that looks like my partner is him, especially if he is wearing a long green coat, even though

I don’t think that my partner’s personal identity overtime is constituted by his green coat: clothing,

like physical appearances, serves as a convenient recognition cue. The heuristic that “seeing is

believing” is, for the most part, convenient and reliable; and it is also automatic (Fiske & Taylor,

1991; Leveroni et al., 2000; Malle, 2004; Shah et al., 2001). Physical cues trigger a constellation of

unique psychological characteristics stored in memory for that person. These processes are so

robust that similar looking faces or voices can trigger them and they often occur despite explicit

knowledge that the person is not the target individual (e.g., Leveroni et al., 2000; Malle, Moses &

Baldwin, 2001; Melzoff & Moore, 1995; Gobbini, Leibenluft, Santiago & Haxby, 2004).

Physical marks, while they are also convenient to detect (i.e., are visible), are also highly

distinguishing (i.e., unique or uncommon). There are situations where the reliance on these types

of physical features is heightened, including where costs of erroneous identification are high, and

when there are multiple contenders for identity. One example of these circumstances is a victim’s

attempt to correctly identify a person who committed a crime among other innocent people;

another, proposed here, is a person’s attempt to identify the return of an ancestor amongst other

non-kin. Thus, we contend that people reason about reincarnation in much the same way they

reason about other people everyday by unconsciously reverting to habitual strategies that enhance

the reliability of their choice. In other words, people are reasoning as empiricists; they

unconsciously weigh up the likelihood of other living person having the same feature (e.g.,

birthmark on the right arm) as another person and the statistical probability that a person at one

time is the same person at another.

Yet why do people around the world seek out distinctive physical marks as evidence of

reincarnation (which assumes that the deceased and living have the same physical properties),

when reincarnation involves a bodily change? Research suggests that these low-level aspects of

perception are not easy to override despite explicit knowledge to the contrary. For example, in

clinical settings, people tend to transfer the psychological characteristics of one individual to

6 Philosophical discussions about the relationship between diachronic and personal identity are relevant but in-depth treatment of them is beyond the scope of this paper (e.g., see Penelhum, 1970 & Shoemaker, 1963).

another on the basis of similar physical characteristics (Addis & Tippet, 2004; Andersen & Cole,

1990; Andersen, Glassman, & Chen, 1995). Conversely, people have difficulty accepting the

dramatic psychological change to individuals who look similar than before the change in cases of

quick-onset of Alzheimer’s disease or an acquired brain injury (Orona, 1999; Ronch, 1996).

These inconsistencies also feature in religious cognition and have been dubbed “theological

incorrectness” (Slone, 2004). Such discrepancies have been observed in people’s reasoning about

other supernatural agent concepts, such as gods (Barrett 1999, Barrett & Keil, 1996; Slone, 2004),

and are explained by the fact that humans are cognitive misers and in certain contexts will revert to

intuitive representations for agents generally (Barrett & Keil 1996; Fiske & Taylor, 1991;

Pyysiäinen, 2004). For instance, Cohen (2007) found that members of an Afro-Brazilian spirit

possession cult tended to attribute the responsibility of a possessed person’s behavior on the

possessed individual, rather than the possessing spirit. For example, if a person was drunk at a

street party and behaved in socially unacceptable ways, even though people said the person was

possessed by a spirit who had displaced the person’s mind, members showed contempt towards the

possessed person for the behavior, and the person would often be sanctioned accordingly. Thus,

people find it difficult to inhibit the association of physical appearances with underlying features

that they associate with underlying personal stability. In the context of reincarnation, people

intuitively revert to the strategies they use to establish the identity of people despite explicit

knowledge that contradicts them; these default intuitions do not simply “shut off” in the face of

explicit ideas about supernatural agents.

Episodic memories indicate underlying personal stability

People are also influenced in their decisions about personal identity by how distinctive a memory

is, especially when this is subject to external verification (for example, that a person remembers

the layout of a historical building). Other research suggests that the inherent qualities of memory

itself also explain why people so often rely upon it in different cultural contexts to establish

reincarnation. Studies have shown that children and adults tend to rely on the stability of

underlying psychological properties, especially memories, to attribute to others a continuous and

numerical identity throughout dramatic physical changes that occur during one’s lifetime and even

in the afterlife (e.g., Astuti & Harris, 2008; Barrett, 1999; Bering, 2002; Bering & Bjorklund,

2004; Bloom, 2004; Boyer, 2001; Cohen & Barrett, 2008, Cohen, Burdett, Knight & Barrett, 2011;

Gutheil & Rosengren, 1996; Newman & Rips, 2005).

Research also suggests that people judge episodic memories (i.e., memories of events that

happened to oneself, such as remembering having dinner the previous night, Tulving, 1972) as

superior to preserving identity, as compared to other psychological features, including other

memory types (e.g., see Blok, Newman & Rips 2005; Klein, German, Cosmides & Gabriel, 2004;

Klein & Nichols 2012; Parfit, 1984; Perry, 1975; Reid, 1969/1785; Shoemaker & Swinburne,

1984). Specifically, in contrast to semantic memory (i.e., knowing that), episodic memory records

events as experienced by the self at a particular time/space (e.g., I remember celebrating my 16th

birthday at my Mother’s house). Like behaviors, they also indicate knowledge about one’s core

concept (i.e., self: e.g., "usually stubborn"), but they also infer properties and relations between the

person at the time of encoding and retrieval: including a self-referential quality of retrieval (i.e., a

sense of “mineness”), that may be the best proxy of underlying psychological stability that people

have access to (Conway, 2005; Klein & Nichols, 2012). Indeed, research on the basis of past life

convictions with contemporary U.S. “spiritual seekers” supports the claim that people regard

episodic memories as convincing evidence of their own past life existence because of the basic

relationship it indicates between them at the time of encoding and recall (White, Kelly & Nichols,

in press). Other research with U.S. adults suggests that these reasoning heuristics also guide

people’s decisions about how to establish continued identity of others following reincarnation. In

what follows, we outline the original studies before turning to the current studies with Jain

participants.

1.2 Previous research on the cognitive foundations of reincarnation

Previous research suggests that people are guided in their decisions about how to establish

continued identity in reincarnation by both the category to which the feature belongs and the

feature’s distinctiveness, and thus, the reliability of evidence. In a series of imaginative

perspective-taking studies conducted by White (in press), U.S. adults completed questionnaires

from the perspective of a village leader, whose job it was to decide, out of a number of potential

people (candidates), how likely it was that they were the reincarnation of the deceased, relative to

others (using a scale, ranging from 1 - Very Unlikely to 5 - Very Likely) and under constrained

conditions (only one candidate could be ranked 5 - Very Likely, 4-Likely and so on). Participants

based their decisions on information about the deceased person’s features, when alive, and the

candidates’ features (each candidate had one similar feature to the deceased). Participants

consistently privileged candidates with a similar episodic memory (i.e., memory of an event) or

physical mark (i.e., a scar) to the deceased as more likely to be the reincarnation over candidates

with other similar features, including: other categorical features (e.g., personality trait, behaviors);

types of memories (e.g., procedural memory); and physical similarities (e.g., physique).

The results of these studies suggest that in addition to the category to which the feature

belonged (e.g., episodic memory or physical mark), the feature’s distinctiveness was also driving

participants’ decisions, especially for physical marks. In one study, for example, participants

ranked the likelihood of candidates with a similar generic and highly distinctive physical mark and

similar generic and highly distinctive episodic memory to the deceased. The examples of the

distinctive features were similar to the generic examples but with additional information that

would make them more distinguishing (e.g., exact size of the physical mark). In this study, more

than one candidate could be ranked in the same position (i.e., more than one candidate could be

ranked as 5 - Very Likely), and thus participants were not as restricted in their choices. The

candidate with a similar distinctive physical mark to the deceased was rated highest overall, and

higher than the candidate with the distinctive episodic memory. Further, the mean scores of the

highly distinctive features were significantly higher than their generic counterparts. Furthermore,

across all studies, these patterns held largely irrespective of whether or not participants believed in

reincarnation, and even though the vast majority explicitly endorsed reincarnation as involving a

bodily change, which suggests that these assumptions are not simply a product of belief or explicit

concepts of reincarnation.

Although the ethnographic research suggests that similar ideas underpin practices around

the world, little is known about whether, and how, these intuitions operate under different cultural

input. Thus, one important task is to establish, under controlled conditions, whether these ideas are

similar or different in another religious tradition, especially beyond western cultures because they

tend to be overrepresented in experimental research (Henrich, Heine & Norenzayan, 2010). These

possibilities were investigated here by presenting Jain adults in Southern India with a modified

version of the imaginative perspective-taking task used in White’s (in press) study.

1.3 Current research with Jain adults in South India

Conducting the study with Jain adults affords a unique and theoretically relevant contrast to

the participants in White’s original study, who were adults in the U.S. with mixed beliefs about the

afterlife. The Jain tradition is reincarnationist and karma determines rebirth. Unlike Buddhism,

however, but like Hinduism (i.e., atman), Jainism endorses the existence of a permanent and

enduring self. The aspect of the self that endures is characterized as jiva, which is often translated

as “soul” but more accurately refers to a “life-monad”. The jiva can undergo constant

modifications and it can inhabit many physical forms (e.g., from an ant to an elephant and even a

plant), what remains the same is often described as pure consciousness, which includes knowledge.

Thus Jainism can be said to be mid-way between western concepts of the self as continuing as

individuals in the afterlife and the Buddhist rendering of the inexistence of the self in life and

rebirth (e.g., see Bering, 2002; Spiro, 1982).

Although Jains accept the existence of a permanent enduring self, what, precisely, the self

entails remains vague. This presents an opportunity to probe folk interpretation of ideas about

reincarnation. Further, like Buddhism, it is thought that with the right prompting, people and other

agents can remember their past life, but this is an exception rather than the norm (see Dundas,

2002). Unlike Buddhism, however, Jainism lacks cultural practices designed to establish the

identity of “special” people (e.g., Lamas) who have been reincarnated and although they accept as

fact that every living person has been, and will be, reborn, Jains are not concerned with

establishing the past life identity of the living (Dundas, 2002; Haraldsson & Samararatne, 1999).

Thus participants are not able to simply draw upon cultural information about how to establish

identity in reincarnation, they are thus more likely to depend upon other input. Since information

about how to identify reincarnated agents is lacking from Jain scriptures it allows for the

possibility that participants will also draw upon intuitive expectations about the stability of

personal identity overtime and how to establish it.

The main aim of this research is to further refine, and test, the extent to which the

aforementioned cognitive biases concerning intuitive expectations of human agents shape such

ideas and corresponding practices in other cultural contexts beyond the western world. Three

questions are of particular interest in the current studies. First, will Jain participants also

distinguish between different types of physical and psychological features in terms of their

likelihood to indicate continued identity in reincarnation? Second, will they privilege physical

marks and episodic memories to the deceased as indicating reincarnation over other features?

Third, to what extent will increasing the distinctiveness of a physical mark and episodic memory

affect participants’ reasoning about how likely they are to indicate a person’s return to the human

world?

Several distinct possibilities follow from what is known about this population. First,

regarding physical features; in Jainism the body is regarded as a vehicle for the jiva but it leaves

the body during the process of rebirth (Dundas, 2002). Thus, one possibility is that participants

will reason according to the doctrine concerning reincarnation and correspondingly, will not regard

physical features – including similar physical marks or even distinctive physical marks – between

the deceased and living as evidence of continued identity in reincarnation. On the other hand, if

participants’ decisions are not governed exclusively by doctrine, then participants will regard

candidates with similar physical features, including distinctive physical marks, as most likely to be

the deceased’s reincarnation than other candidates. Second, regarding psychological features; one

possibility is that participants will not judge candidates with similar memories to the deceased as

more likely to the deceased’s reincarnation than candidates with other similar psychological

features because Jainism does not endorse the view that ordinary people can remember their past

lives. Another possibility is that Jain participants will privilege episodic memories, and especially

distinctive episodic memories, as evidence for continued identity.

A series of studies were conducted to explore these possibilities. Ethical approval for all

studies was obtained from the School Research Ethics Committee (History and Anthropology),

Queen’s University, Belfast. In the current research, participants were selected from the university of

Madras, Chennai, the capital of the Indian state of Tamil Nadu and the forth largest city in India – with

a population of around 5.9 million. Jains make up approximately and 13% of Tamil Nadu (and 4% of

the population of India), where they are a minority religion (Bose, Singh, & Adhikary, 1991). The

participant sample included two sects of Jainism, both the Digambaras and Svetambaras. While both

have unique histories, development, and conflicts, they share many of the same basic principles of

Jainism, differing in terms of features and practicalities of the religion, such as whether women can

attain enlightenment and whether their ascetics are naked or dress in white robes (Dundas, 2002). The

university of Madras attracts many Jain students due to the Jainology department and active

recruitment and promotion of Jainism. In the study, demographic information was collected but all

surveys were anonymous. All participants were enrolled adult English-speaking students at the

university and identified as belonging to the Jain religious tradition. All volunteered, without

compensation, to participate in the study.

The studies took place on, or close to, the University of Madras. Paper questionnaires were

administered to participants either by the lead researcher or by a paid faculty member at the

university, who served as a research assistant. Participants completed the questionnaire, typically

individually or in small groups of around four students under the supervision of the researcher. For all

studies, participants were given written instructions; including that they would be asked to complete a

questionnaire from the perspective of another Jain and to answer the questions as quickly as possible,

without giving them too much thought. Participants gave informed written consent before proceeding

to the main study task and were given an opportunity to ask questions about the study. Given that

karma is the central organizing system of rebirth, all participants were verbally instructed that there

was no information on how good or bad the deceased or living candidates were. The study design

differed from the original study with U.S. participants (White, in press) because it was within-

subjects for studies one and two. This departure from the original design was pragmatic, given the

difficulty in obtaining adequate numbers of participants in a cross-cultural setting. Thus, the same

participants completed studies one and two; in the same questionnaire, and in the order they are

presented here. Study 3 was conducted after the first two studies and a new set of participants

completed the study.

In the main studies, like the U.S. version, participants were presented with a series of

perspective-taking tasks in which they played the role of a Jain village leader, whose job it was to

rank, out of a number of potential candidates, how likely it was that particular candidates were

reincarnation of the deceased, relative to other candidates7. The aim of the first study was to

determine whether Jain participants in South India would select similar features to those used in

identification procedures in reincarnation contexts as documented by ethnographic reports, and in

particular, to U.S. adults with mixed afterlife beliefs as determined by White’s (in press) study. Like

the original study, in order to maximize differences between candidates, and to ensure participants

used the full spectrum of ranks, each candidate had to be placed in a different rank. For example, only

one candidate could be ranked 5 - Very Likely, 4-Likely and so on. If participants reasoned similarly

to U.S. participants, namely, that candidates with a similar episodic memory and physical mark were

most likely to the deceased’s reincarnation, then the mean rank scores for candidates with a similar

episodic memory or physical mark to the deceased would be significantly higher than all other

candidates.

2. Study 1

2.1 Method

Participants

Sixty-three adults participated in the study. Participants were students at the university of Madras

(mean age = 22 years, range 18 – 31 years). Of these, 64% were female and 36% male. The

majority of the participants belonged to the Svetambara sect (84%) and a minority belonged to the

Digambara sect (16%).

Materials and procedure

Participants were presented with the following information:

Imagine that you are the leader of a Jain village where they believe that everyone who dies

is reborn. As the leader, you must decide how likely five people are to be the reincarnation

of the deceased, based on information about the deceased, when he was alive, and these 7 The studies were designed as perspective-taking tasks to prevent U.S. participants, who did not believe in reincarnation, from misinterpreting the task as based upon their actual belief in the afterlife, rather than how they reason about scenarios concerning it.

individuals. You must rank them in terms of the likelihood that they are the reincarnation

of the deceased. You will now read a description of the deceased person, when he was

alive, and decide how likely it is that each of individuals presented is the reincarnation.

There is no right or wrong answer. There are no tricks. You have to make an informed

judgment based on the information you have.

Participants then completed a perspective-taking task, where they read information about the

deceased person when he was alive8 and information about five living people who were potentially

the reincarnation of the deceased. In the study, participants had to rank the five candidates in order

of likelihood that they were the reincarnation of the deceased, using a forced-choice 5-point scale,

ranging from 1 – Very Unlikely to 5 – Very Likely. Like the U.S. version, the deceased was

described as having five features. The main question of interest was whether any of the five

candidates would be ranked consistently higher or lower than the others, especially candidates with

a similar PHYSICAL MARK and an EPISODIC MEMORY to the deceased. The other features

included were the same as those in the original U.S. study and were selected because they appear

in the ethnographic records as guiding people’s decisions in real world contexts. These features

were: a characteristic BEHAVIOR and PERSONALITY trait and again a fifth property, NAME,

was included to provide an example of a similarity that was not dependent upon physical or

psychological properties.

Participants were presented with similar examples of features as the original study. In the

original U.S. version, six research assistants, who were blind to the study hypotheses, selected the

examples for all studies from a longer list. Specifically, the examples in Study 1 and 2 were

selected on how “neutral” they were (e.g., did not give unnecessary additional information to the

participants other than the feature, such as a memory that would cause an emotional reaction). The

examples were also matched by the research assistants on how likely they would be for a “normal”

8 In order to control for the possible influence of gender on participants’ judgments, which is especially pertinent given the controversy over whether women can attain enlightenment in Jain sects (e.g., see Dundas, 2002, p.55), the deceased and the five living people were described as a male in both tasks.

person to exhibit (i.e., did not indicate something extraordinary about the person), and most of

these examples were included in the current study. Some of the items in the examples were,

however, modified to make the vignettes convincing and standardize, to the extent possible, the

likelihood of Jains in South India possessing such features or sharing the experiences given.

Specifically, the item described in Study 1 (a tennis racquet) was modified to become a more

culturally familiar object (a ball) and the behavior was changed from the candidate nodding his

head a lot when he spoke (which is common in South India) to the candidate widened his eyes a lot

when he spoke9. The names of the candidates were also modified from those common in America

to those common in India. Further, the names of the candidates were selected so that they did not

indicate caste, economic status or have an association with Jain historical figures. Like the U.S.

version of the study, as a precaution against the possibility that something about the particular

example other than the type of feature given was determining participants’ judgments, participants

ranked candidates in two tasks with different examples of the features given in each task.

Perspective taking tasks

In Task 1, participants read that

The deceased was called Akash [NAME10]. He had a mole on his left arm [PHYSICAL

MARK]. He widened his eyes a lot when he spoke [BEHAVIOR]. He remembered the day he

found a ball [EPISODIC MEMORY]. He was shy when he met new people [PERSONALITY].

The deceased was called Salil [NAME]. He had a scar on his right leg [PHYSICAL

MARK]. He used his hands a lot when he spoke [BEHAVIOR]. He remembered the day he found

a pair of shoes [EPISODIC MEMORY]. He was outgoing and liked to meet new people

9 Note that the western coders did not rate the examples used in this version of the study with Jain participants in terms of how likely they would be for a normal person to exhibit. The differences in this study, however, compared to the U.S. version are minor. 10 For all studies, parenthetical notes [i.e. EPISODIC MEMORY] were not included in the survey but are provided here for clarification.

[PERSONALITY].

Participants then placed, in rank order, five candidates. Each candidate had one matching feature to

the deceased. For example, in Task 2, candidates were described as the following

Candidate 1: Was called Salil.

Candidate 2: Had a scar on his right leg.

Candidate 3: Used his hands a lot when he spoke.

Candidate 4: Remembered the day he found a pair of shoes.

Candidate 5: Was outgoing and liked to meet new people.

Participants then ranked each of the five candidates depending on how likely they were, in relation

to the other candidates, to be the reincarnation of the deceased.

2.2 Results

Mean and mode rank scores for Tasks 1 and 2

Mean scores for each feature type in Tasks 1 and 2 are presented in Table 111. Figure 1 also

displays the mode ranking across both tasks, which indicates the most frequent rank for each

feature. Of particular interest, the frequencies show that EPISODIC MEMORY and PHYSICAL

MARK categories are clearly privileged as cues to continued identity in reincarnation across both

tasks, which constituted 126 placements. First, the candidate with the similar EPISODIC

MEMORY to the deceased was most often ranked 5 – Very Likely in both tasks than any other

rank. Consider some selected frequencies in addition to those displayed in Table 1. Candidates

with a similar EPISODIC MEMORY were placed in the highest possible rank category 73% (n =

92) of the time, and the second highest rank category (4 – Likely) 19.8% (n= 25) of the time.

Second, participants ranked the candidate with the similar PHYSICAL MARK to the deceased as 4

– Likely in 61.9% (n = 78) of all decisions, and as 5 – Very Likely in 22.2% (n = 28). Overall, at

least 84% of all judgements for candidates with a similar EPISODIC MEMORY or PHYSICAL

11 Preliminary analyses showed no significant effect of gender or sect on mean scores and they were therefore were collapsed in all subsequent analyses and are not reported in participant demographics for Study 3.

MARK ranked them in the highest possible two categories. This demonstrates a consistent

privileging of these categories as indicators of continued identity. By comparison, the distribution

of rank scores across other categories was less homogeneous. For example, the candidate with the

similar PERSONALITY to the deceased was ranked 3 – Neither Likely nor Unlikely in 46% (n =

58) of all cases, but also 2 – Unlikely in 38.1% (n = 48) and 1 – Very Unlikely in 14.3% (n = 18).

Assessing Task Validity: Comparing Mean Rank Scores for Individual Categories in Tasks 1 and 2

To assess task validity, five paired-samples t-tests of the mean scores for five categories in Tasks 1

and 2 were conducted. There was no significant difference between the five category examples in

the two tasks [EPISODIC MEMORY, Task 1 (M = 4.86, SD = 0.66), Task 2 (M = 4.46, SD =

0.66), t(62) = 3.88, p = 0.69.; PHYSICAL MARK, Task 1 (M = 4.25, SD = 1.78), Task 2 (M =

3.87, SD = 1.75), t(62) = 3.38 , p = 0.72; PERSONALITY, Task 1 (M = 1.97, SD = 1.72), Task 2

(M = 2.76, SD = 1.74), t(62), = 5.08 , p = 0.59; BEHAVIOR, Task 1 (M = 1.84, SD = 1.52), Task 2

(M = 1.83, SD = 1.51), t(62), = 2.49, p = 0.15; NAME, Task 1 (M = 1.54, SD = 1.61), Task 2 (M =

1.68, SD = 1.58), t(62), = 6.46, p = 0.52]. Therefore, participants reasoned similarly about the type

of feature in Tasks 1 and 2.

Table 1: mean score and mode rank for features in Tasks 1 and 2

Task 1 Category and examples

Mean Score SD Mode Task 2

Category and examples Mean Score SD Mode

Episodic memory (finding a ball) 4.86 0.66 5 Episodic memory

(finding a pair of shoes) 4.46 0.66 5

Physical mark (mole on left arm) 4.25 1.78 4 Physical mark

(scar on right leg) 3.87 1.78 4

Personality (was shy) 4.25 1.72 2 Personality

(was outgoing) 2.76 1.72 2

Behavior (widened eyes when spoke)

1.84 1.52 3 Behavior (used hands when spoke) 1.83 1.52 3

Name (Akash) 1.54 1.61 1 Name

(Sali) 1.68 1.61 1

Figure 1: mode rank for features in Tasks 1 and 2 combined

Table 2: mean and mode rank for features in Tasks 1 and 2 combined

Category Mean rank Mode rank

Episodic memory

4.73a 5

Physical mark

4.15a 4

Personality

2.54a 3

Behavior

1.95 2

1.64 1

Note: Means with superscripts are significantly higher than means below them in the same column.

Mode Rank

Episodic memory

Physical Mark

Personality

Behavior

Statistical analyses: comparing mean rank scores for all categories

As there were no significant differences between mean scores in the five category examples (e.g.,

EPISODIC MEMORY, PHYSICAL MARK, PERSONALITY etc.) across the two tasks, data for

Tasks 1 and 2 were combined and analyses were conducted on the mean rank scores for each

category (see Table 2). A Friedman’s test was conducted to determine whether there was a

difference in the mean rank scores for each of the five candidates, and thus, whether there was an

uneven distribution of ranks for each category type. There was a statistically significant difference

in mean rank score for the candidates depending on the category to which the similar feature

belonged, χ2(406.4) = 7.600, p = 0.01. Wilcoxon signed-rank tests showed that most of the

category features elicited a statistically significant increase in mean rank, namely, from

BEHAVIOR to PERSONALITY (Z = -5.116, p = 0.000), PERSONALITY to PHYSICAL MARK

(Z = -9.323, p = 0.00), and PHYSICAL MARK to EPISODIC MEMORY (Z = -6.808, p = 0.00),

but not between NAME and BEHAVIOR (Z = -2.656, p = 0.008). Therefore, PHYSICAL MARK

was ranked significantly higher than all other candidates including EPISODIC MEMORY, and

EPISODIC MEMORY was ranked significantly higher than all other candidates except

PHYISCAL MARK.

2.3 Discussion

Jain participants discriminated between the likelihood of living candidates to be the deceased’s

reincarnation based upon a single similar feature between the deceased and living. They

consistently ranked two features, episodic memory and especially physical marks, as the most

reliable cues to continued personal identity among other types. These two features were also

privileged as indicators of continued identity in reincarnation by participants in the U.S., who

received a modified version of this task, and to reports of how people establish reincarnation

around the world. These results are significant in two ways.

First, it is noteworthy that participants ranked physical marks so highly as indicating

continued identity, given that the body changes following rebirth in Jain theology. One

interpretation of this finding is that the participants do not actually represent reincarnation as

entailing a bodily change and this concept has been erroneously superimposed upon them. To

investigate this, in Study 2, participants were asked about their idea of what reincarnation entails.

Another, more plausible interpretation of this finding is that Jain participants are also influenced in

their decisions about reincarnation by unconscious processes of person recognition used to identify

people on a daily basis everywhere. As discussed, these assumptions that “seeing is believing” are

not easily inhibited by explicit beliefs to the contrary. A further question of interest is whether Jain

participants, like their U.S. counterparts, would distinguish between the likelihood of different

types of physical features to indicate continued identity in reincarnation. This question was

investigated in Study 2 by asking participants to rank candidates with different types of physical

features, including basic facial similarities or bodily marks, to be the deceased’s reincarnation.

The second finding of interest is that the candidate with the similar episodic memory had a

significantly higher mean rank score than all other candidates, including the candidate with the

similar physical mark, although the frequencies indicate that both features are regarded as robust.

Thus, the question then becomes why do Jain participants regard memory as such a robust

indicator of continued identity after reincarnation? One possibility, as discussed in the

introduction, is that participants regard memory as something that provides evidence of continued

identity because of the unique psychological relationship it conveys between the encoder and

retriever. Although participants may be reasoning that memory provides evidence of jiva (i.e.,

consciousness), Jain doctrine does not discriminate between different types of memory. For

example, a memory of a fact (i.e., SEMANTIC) should provide equally robust evidence of

consciousness as a memory of an event. One way to investigate this is to assess whether

participants will, like U.S. participants, privilege episodic memories as identity markers over other

memory types. These questions were investigated in Study 2.

3. Study 2

3.1 Method

The same participants in Study 1 completed another two perspective-taking tasks in this study. In

Task 1, participants had to rank five candidates in terms of how likely they were to be the

reincarnation of the deceased, again, only one candidate could be placed in each rank. To

determine whether participants privileged physical marks over other physical features, candidates

had a different corresponding physical feature to the deceased when he was alive, namely: a

physical mark, facial feature, physique, physical abnormality and hair type (see Table 3). In Task

2, participants ranked five candidates with different types of memories in terms of their likelihood

to be the reincarnation of the deceased (see Table 3). Three sub-types of long-term memory (i.e.,

declarative, procedural, semantic) and one memory ability (i.e., aptitude) were included to

investigate whether episodic memories were perceived as more robust cues to continued identity

than other kinds of memory.

Perspective-taking tasks

The deceased was called Ebi. He had a patch of skin discoloration on his leg [PHYSICAL

MARK12]. He had a long pointy hooked nose [FACIAL]. He was tall with a hunched back

[PHYSIQUE]. He limped due to a flat foot [ABNORMALITY]. He had thick, curly multicolored

hair [HAIR].

The deceased was called Javad. He remembered seeing a shoe for the first time [EPISODIC

MEMORY]. He was good at remembering people’s names [APTITUDE]. He remembered the

names of shoe parts [SEMANTIC]. He remembered how to mend shoes [PROCEDURAL]. He

remembered recently learned facts about shoes [DECLARATIVE].

Participants’ concept of reincarnation

Participants selected the statement that they thought most accurately characterized reincarnation

from a list of four options. The purpose of this question was to determine whether participants held

a concept of reincarnation as entailing bodily change. The expectation was that most participants

would select option three, the statement most closely corresponding to the Jain theological

characterization of reincarnation (Matlock & Mills, 1994; Obeyeskere, 2002). To reduce potential

bias in the study and afterlife questionnaire, participants were presented with this question at the

end of the study. The options below were:

1. …the body and the person permanently ceases to exist” [extinctivist13]

2. …the body and the person continue to exist forever” [immortalist]

3. …the body ceases to exist and the person continues to exist in a new body”

[reincarnationist]

4. …the body continues to exist and the person ceases to exist” [eclectic]

3.2 Results

Mean and Mode Rank Scores

As displayed by Table 3, candidates with a similar PHYSICAL MARK and EPISODIC

MEMORY to the deceased had a higher mean rank than other candidates in their respective

categories, and both were ranked most often as 5 – Very Likely. Additional frequencies confirm

that both were ranked consistently highly. For PHYSICAL MARK 60.3% (n = 38) of participants

ranked the candidate as 5 – Very Likely and 25.4% (n = 16) as 4 – Likely. The frequencies for

EPISODIC MEMORY likewise suggest that participants’ also ranked the candidate consistently

highly: 74.6% (n= 47) selected 5 – Very Likely and 14.3% (n = 9) selected 4 – Likely. Therefore,

for both categories, of all ranks (1-5) possible, at least 85% of all participants placed them in the

highest two ranks (4-5).

Statistical analyses: comparing mean rank scores for PHYSICAL features

A Friedman’s test was conducted to determine whether there was a difference in the mean rank

scores for each of the five candidates in the PHYSICAL category. Results revealed an uneven

distribution of ranks depending upon the feature type χ2(4) = 183.5, p = 0.00. A series of Wilcoxon

signed-rank tests were conducted to pinpoint where the significance lay. Results showed a

significant increase in most mean rankings. These were: from PHYSIQUE to FACIAL (Z = -

5.125, p = 0.00), FACIAL to ABNORMALITY (Z = -6.228, p = 0.00) and ABNORMALITY to

PHYSICAL MARK (Z = -4.384, p = 0.00), but not from HAIR to PHYSIQUE (Z = -1.135, p =

0.25). Of particular interest was the finding that the candidate with the similar PHYSICAL MARK

to the deceased was ranked significantly higher than all other candidates.

Statistical analyses: comparing mean rank scores for MEMORY features

Another Friedman’s test was conducted to determine whether any the mean rankings for the five

candidates in the MEMORY category differed significantly. There was a statistically significant

uneven distribution of ranks depending upon feature type, χ2(4) = 211.3, p = 0.00. A series of

Wilcoxon signed-rank tests demonstrated a significant increased in two mean rankings: from

PROCEDURAL to SEMANTIC (Z = -7.077, p = 0.00), SEMANTIC to EPISODIC (Z = -5.374, p

= 0.00), but not from APTITUDE to DECLARATIVE (Z = -2.285, p = 0.22) or from

DECLARATIVE to PROCEDURAL (Z = -1.026, p = 0.22). Of most importance, the candidate

with the similar EPISODIC MEMORY to the deceased was ranked significantly higher than all

other candidates.

Table 3: mean and mode rank for features in Tasks 1 and 2

Task 1 Physical category: features and examples

Mean rank

Task 2 memory category: features and examples

Mean rank

Physical mark (patch of skin discoloration on leg)

Episodic memory (seeing a shoe for the first time)

Abnormality (limped due to a flat foot)

Semantic (names of shoe parts)

Facial feature (long pointy hooked nose)

Procedural (how to mend shoes)

Physique (tall with a hunched back)

Declarative (recently learned facts about shoes)

Hair (thick, curly multi-colored hair)

Aptitude (good at remembering people’s names)

Note: Means with superscripts are significantly higher than means below them in the same column with the corresponding superscript.

3.3 Discussion

The results of Study 2 reinforce the results of Study 1 by showing that Jain participants

privileged particular types of physical and mnemonic cues to establish continued identity in

reincarnation, especially: physical marks and episodic memories. In addition, Study 2 also

demonstrated that Jain participants privileged physical marks and episodic memories over other

physical and mnemonic features. Again, these findings parallel findings from the U.S. version of

this task with participants who held mixed views about the fate of the body after death, and cross-

cultural reports, which suggest that these two types of features are frequently used to determine the

identity of reincarnated agents.

These results beg the question of why episodic memories and physical marks are privileged

over other, similar physical and mnemonic features. The consistent use of episodic memory by

Jain participants requires explanation. The one proposed here is that they implicitly represent

episodic memory as a robust indicator of continued identity because of the basic psychological

relationship it conveys between the encoder and retriever. Another finding that necessitates further

attention is the use of physical marks to establish identity in reincarnation. This is of particular

interest because Jain dogma presents the physical body as changed in reincarnation. This study

also confirmed that the vast majority of participants endorsed a concept of rebirth that aligned with

Jain doctrine, ruling out the possibility that we were superimposing a western concept of

reincarnation, which entailed bodily change, upon them. Thus, on the one hand, participants

claimed that the body changed in the process of rebirth but on the other, reasoned as though

physical cues, and especially physical marks, were robust indices to continued personal identity in

these contexts. This is commensurate with the theory proposed in the introduction; explicit beliefs

about personal identity do not inhibit mundane processes of person recognition where physical

similarities provide reliable evidence of continued identity.

It is of further interest, however, because other physical features (e.g., a peculiar shaped

nose) likewise depend upon convenient recognition cues. If participants do reason in terms of the

mundane processes of personal recognition, then why do they privilege physical marks in

particular? There are likely to be many factors that influence participants’ decisions in the tasks

that are not dependent on the type of feature alone. For instance, participants may be acutely aware

of any features that indicate an unusual occurrence, how unusual they are for a “normal” person to

exhibit, whether the features are present from birth and even the probability of surface features

being alike (e.g., similar name). Yet in most of the tasks, these individual factors do not appear to

sway participants’ judgments on their own, rather, certain types of features are consistently

privileged even though different examples of the features are given. It may be the case that some

features are regarded as stronger cues to continued identity when accompanied by some of these

additional factors.

One factor hypothesized to elicit this effect is the distinctiveness of the features. As

discussed, White (in press) demonstrated that U.S. participants were influenced in this task

according to how distinguishing the feature was. Thus, in situations requiring the identification of

persons, participants may unconsciously revert to habitual strategies they use to establish identity

everyday by seeking out physical similarities between people at two times. One possibility is that

in the current task Jain participants were reasoning about these features in terms of their

distinctiveness, in other words, how unique or uncommon they would be in the general population.

Perhaps, for instance, people regard bodily marks as highly distinctive. Likewise, in addition to the

assumption that episodic memory entails personal ownership over the memory, perhaps

participants also implicitly represent episodic memories as containing more detailed and specific

information than other memories, and they are reasoning also according to the distinctiveness of

the memory. Furthermore, one limitation of the previous studies was that participants were forced

to rank each candidate in a different rank position, so that only one could be 5 - Very Likely, 4 -

Likely and so on, because this limited the freedom of participants to rank several candidates as

equally likely or unlikely. Therefore, in the final study, like the U.S. version, participants could

rank multiple candidates in the same rank order, which gives a more reliable indication of the

reliability assigned to different candidates, independent of others.

The aim of the third, and final study with a new set of participants was to investigate the

extent to which the distinctiveness of a feature (i.e., how distinguishing or unique), and/or the

category to which it belonged (i.e., PHYSICAL MARK or EPISODIC MEMORY), was driving

participants’ decisions. This was investigated by manipulating the distinctiveness of the

PHYSICAL MARK and EPISODIC MEMORY in the task, so that the task contained one generic

and distinctive PHYSICAL MARK, and one generic and distinctive EPISODIC MEMORY. There

were three main questions of particular interest guiding this study. First was whether participants

would rank one generic example higher than the other in the respective category (i.e., PHYSICAL

MARK or EPISODIC MEMORY), which would indicate that examples were privileged on the

basis of the category to which they belonged. There are two distinct possibilities. On the one hand,

participants could reason as though one of the candidates with a similar generic feature was more

likely to be the deceased’s reincarnation because that feature provides better evidence of it. On the

other hand, they may regard both generic examples highly and more-or-less equally likely to

indicate continued identity and, like their U.S. counterparts, rank both generic features similarly.

Second was the question of whether there would be a significant increase between the

mean rank score of the candidates within the categories (i.e., from generic to DISTINCTIVE

PHYSICAL MARK and generic to DISTINCTIVE EPISODIC MEMORY), which would indicate

a role for distinctiveness instead of, or in addition to, the category to which the feature belonged.

Following the results of the U.S. version of the study, and based upon previous research, which

suggests that distinctiveness plays a role in both but especially in memory, again there are two

likely possibilities. One possibility is that there would be a significant increase in the rank from

generic to distinctiveness in both categories. Another possibility, which is based upon findings

with U.S. participants, is that this trend would be significant for the PHYSICAL MARK only.

The third, and final question explored in Study 3 was whether there would be a significant

difference between the mean rank score of the two distinctive features, namely, between

DISTINCTIVE PHYSICAL MARK and DISTINCTIVE EPISODIC MEMORY. Although this

question was not statistically addressed in the U.S. version of the study, mean scores between the

two were remarkably similar (0.35 difference), which suggests that there would have been no

significant difference between the two. If Jain participants reason similarly to their U.S.

counterparts and judge the combination of feature and distinctiveness as equally compelling, then

there would be no significant difference between the mean rank score. If, on the other hand, Jain

participants regard the combination of one (e.g., EPISODIC MEMORY plus DISTINCTIVE) as

more compelling than another (e.g., PHYSICAL MARK plus DISTINCTIVE), then there would

be a significant difference between mean rank scores of the two distinctive features.

4. Study 3

4.1 Method

Participants

Thirty-four new participants completed the study. Participants were students at the university of

Madras (mean age = 19 years, range 18 – 32 years). Of these, 89% were female and 11% male.

They completed pencil-and-paper questionnaires administered by a research assistant.

In this study, participants had to rank four candidates in terms of the likelihood that they were the

deceased’s reincarnation on a scale from 1 - Very Unlikely to 5 - Very Likely. More than one

candidate could be ranked in the same position (i.e., more than one candidate could be ranked as 5

- Very Likely, 4 - Likely and so on). The deceased was described in the tasks as having four

characteristics when alive: two types of physical marks (a mole and a birthmark) and two

memories: one memory of receiving a pair of shoes and another memory of receiving a ball. For

each of the categories (i.e., PHYSICAL MARK and EPISODIC MEMORY), one of the examples

was comparatively lower (i.e., generic) than the other in terms of distinctiveness. Again, six

volunteer assistants (prior to the original, U.S. version of the study), who were blind to the study

hypotheses, determined the features’ distinctiveness as high or generic.

The examples of the PHYSICAL and EPISODIC MEMORY in the task were similar, but

one of them contained additional information that would make them more distinctive (i.e., unique

or unusual and thus statistically improbable for other people to have). For example, a 2.5cm

birthmark on a person’s right arm is highly distinctive physical mark, whereas a mole on a

person’s left arm is less distinctive and thus, generic. More people have moles on their arms than

people have 2.5cm birthmarks. Therefore, in the task, there was one highly distinctive and generic

EPISODIC MEMORY one highly distinctive and generic PHYSICAL MARK. To avoid potential

effects of presentation order upon the participants’ decisions, the distinctive and generic examples

were randomized so that paper copies of the questionnaire contained examples in different order

(i.e., distinctive or generic example first).

Perspective taking tasks

Participants read the following information

The deceased was called Akash. He had a mole on his left arm from birth [PHYSICAL

MARK]. He had a 2.5cm birthmark on his right arm [DISTINCTIVE PHYSICAL MARK]. He

remembered the day he received a new pair of black shoes [EPISODIC MEMORY]. He

remembered the day he received a green sports ball with a 1.5cm hole in it [DISTINCTIVE

EPISODIC MEMORY].

4.2 Results

Mean and mode rank scores

Selected frequencies (including mean rank, mode data and a breakdown of percentages for the

highest rank) are displayed in Table 4. The candidate with a similar DISTINCTIVE PHYSICAL

MARK to the deceased was ranked on average highest overall (M = 3.32). Three candidates

(DISTINCTIVE PHYSICAL MARK, PHYSICAL MARK and DISTINCTIVE EPISODIC

MEMORY) were most frequently assigned the highest possible rank (5 – Very Likely) and one

candidate, with a similar (EPISODIC MEMORY), was most often assigned the second highest

possible rank (4 – Likely). All four candidates were ranked consistently highly. For example, for

each of the four candidates, between 32.4% (n = 11) and 85.3% (n = 29) of participants assigned

them the highest possible rank (5 – Very Likely). More generally, participants tended to assign

candidates the higher ranks (4 and 5) and seldom assigned any candidates to the lower ranks (1, 2

and 3). For instance, for all four candidates, at least 73% (n = 99) of the total ranks assigned (n

=136) were 4 or 5. Finally, as expected, candidates with the highly distinctive version of the

features (i.e., DISTINCTIVE PHYSICAL MARK, DISTINCTIVE EPISODIC MEMORY) were

ranked on average higher than their generic counterparts (i.e., PHYSICAL MARK, EPISODIC

MEMORY).

Statistical analyses: comparing mean rank scores

A Friedman’s test was conducted to determine whether the difference in mean rankings for the

candidates were significant. There was a statistically significant difference in mean rank score for

the candidates, χ2 (3) = 26.3, p = 0.00. A series of four Wilcoxon signed-rank tests were conducted

to explore where the difference lay. Of particular interest was whether the candidates differed

significantly according to: the category to which they belonged (generic PHYSICAL MARK or

EPISODIC MEMORY); the distinctiveness of the feature across categories (i.e., from generic

PHYSICAL MARK to DISTINCTIVE PHYSICAL MARK and generic EPISODIC MEMORY to

DISTINCTIVE episodic memory) and within the categories (i.e., DISTINCTIVE PHYSICAL

MARK or DISTINCTIVE EPSIODIC MEMORY).

The mean rank score for the generic EPISODIC MEMORY was not significantly higher

than the generic PHYSICAL MARK, (Z = -.738, p = 0.46). There was a significant increase in the

mean ranking within categories and by distinctiveness from PHYSICAL MARK to

DISTINCTIVE PHYSICAL MARK (Z = -3.898, p = 0.00), and from EPISODIC MEMORY to

DISTINCTIVE EPISODIC MEMORY (Z = -2.133, p = 0.03). Further, there was no significant

increase in the mean rank scores for distinctive candidates across categories, from the

DISTINCTIVE PHYSICAL MARK to the DISTINCTIVE EPISODIC MEMORY (Z = -1.453, p =

0.06).

Table 4: Mean and mode rank for features

Features and examples

Mean rank

N ranked as 5 (Highly Likely)

% ranked as 5 (Highly Likely)

Distinctive episodic memory (receiving a pair of black shoes)

2.41 b

Distinctive physical mark (2.5cm birthmark on right arm)

3.32 a

Generic episodic memory (receiving a green sports ball with 1.5cm hole in it )

2.19 b

Generic physical mark (mole on left arm from birth)

2.07 a

Note: Means with superscripts are significantly higher than means below them in the same column with the corresponding superscript.

4.3 Discussion

In general, the results of Study 3 fit the theory that mundane processes of identification, especially

considerations about the statistical probability of a person sharing a particular feature with another

person, affect reasoning about personal continuity in reincarnation. The results also showcase the

cross-cultural similarities in reasoning about reincarnation between Jain participants in South Asia

and U.S. participants. Three findings are of particular interest.

First, the data suggest that Jain participants regard physical marks and episodic memories

as more-or-less equally robust cues to establish continued identity in reincarnation because the

candidates were not ranked differently according to the category to which the features belonged

(generic PHYSICAL MARK or EPISODIC MEMORY). This finding reinforces the findings of

Study 1 and 2. Even though Study 1 suggests that the candidate with a similar physical mark was

ranked significantly higher than the candidate with a similar episodic memory to the deceased, the

more general pattern of findings, and especially the frequencies across all studies, show that both

features are regarded as robust cues to continued identity. Thus, the findings should not be

interpreted as indicating that these categories are not important; on the contrary, there is a ceiling

effect reported in the frequencies across studies 1-3 (i.e., most participants ranked all candidates as

5 – Very Likely or 4 - Likely), which indicates that both physical marks and episodic memories are

consistently categorically privileged as cues to continued identity. These findings also parallel

those with U.S. participants who also regarded both generic examples of these features similar in

terms of their likelihood to indicate continued identity.

The second finding of interest is the significant increase in the mean rank score from

generic to DISTINCTIVE PHYSICAL MARK and generic to DISTINCTIVE EPISODIC

MEMORY. This suggests that in addition to the high baseline by virtue of candidates possessing a

similar physical mark or episodic memory to the deceased, the distinctiveness of these features

increased the strength of participants’ judgments that they indicate continued identity in

reincarnation. This result provides a clear confirmation that distinctiveness of features influence

participants’ judgments, even when this contradicts explicit ideas about reincarnation as involving

a bodily change. This finding can be contrasted to the U.S. version of the study, where the increase

from rank scores between generic and distinctive features were significant for the physical mark

only. It is important not to overestimate the extent of these differences however: both U.S. and Jain

participants’ rankings increased for both features as a product of distinctiveness, the difference lies

in the extent to which they increased for episodic memory.

The third, and final finding of interest is the similarity in mean rank score for both the

DISTINCTIVE PHYSICAL MARK and the DISTINCTIVE EPISODIC MEMORY, and the

corresponding non-significance in magnitude between them (p = .06). This suggests, but does not

offer conclusive evidence for, the possibility that Jain participants, like their U.S. counterparts,

regard both feature combinations (i.e., physical marks plus distinctiveness and episodic memory

plus distinctiveness) as more or less equally robust cues to continued identity in reincarnation.

5. General discussion

Cross-cultural reports and studies with U.S. participants suggest that people in different

contexts regard the perceived continuity of episodic memories and physical marks as providing

evidence of continued personal identity in reincarnation. The aim of this research was threefold:

first, to determine whether people in South Asia, from the reincarnationist tradition of Jainism,

differentiate between different types of physical and psychological features as cues to continued

identity; second, to establish whether they also show a preference for episodic memories and

physical marks and third; to understand the extent to which feature distinctiveness influences their

decisions. The findings suggest that in all three questions, they do. Jain participants, who explicitly

endorsed a concept of reincarnation where the physical body changes but where consciousness

continues, consistently privileged physical marks and episodic memories over other personal

characteristics and especially those that were highly distinctive.

Perhaps the most empirically robust findings concern the influence of feature

distinctiveness on physical marks and episodic memories, as reported in study 3. In this study,

unlike studies 1 and 2, participants were able to use any of the five rank options for all features. By

comparison, in studies 1 and 2, participants could only select one rank per candidate (e.g., only one

could be ranked as 5 - Very Likely) and one weakness of the study design in the first two studies is

that participants did not have the freedom to independently rank the utility of each of the features

as cues to identifying reincarnated agents. It may be the case, therefore, that if given freedom

participants may have ranked some other features (e.g., personality trait, semantic memory) as

higher than the current design permitted. Therefore, the current study may underestimate the extent

to which these other features provide evidence of continued identity in reincarnation.

Likewise, the findings concerning the role of physical marks in identity judgments are

more conclusive than those concerning episodic memories. We have identified some relevant

cognitive constraints to explain similarities in reasoning about physical marks as evidence of

reincarnation, especially heuristics concerning person recognition, such that “seeing is believing”

and the statistical probability that a feature occurs in two people, and thus an index of the feature’s

reliability. We have, however, also proposed, but not tested, the possibility that decisions

concerning episodic memory may also be underpinned by universal assumptions about what

indicates underlying personal stability. Determining which properties of episodic memories led

people to use them to establish personal identity in reincarnation is worthy of further investigation.

Despite these limitations, the results of this research are significant in several ways. First,

they complement the ethnographic literature by demonstrating, under controlled conditions, the

relative invariance of reasoning about how to identify reincarnated agents. Jain adults from South

Asia and U.S. adults who held mixed beliefs about the fate of the person after death, reasoned

similarly in the tasks. Namely, they reasoned as though similar physical marks and episodic

memories between the deceased and living indicated that they were one and the same person

through the process of reincarnation. These similarities held even when it contradicted theological

doctrine, such as the divergence between explicit beliefs about reincarnation as involving a bodily

change and yet regarding physical marks as a robust indicator of it. Thus, Jains, like people

everywhere who use physical marks to establish identity, reason about reincarnation in

theologically incorrect ways.

Second, these studies also demonstrate that ideas about reincarnation, like other “religious”

concepts, are grounded in ordinary cognition and subject to cognitive constraints (Barrett, 2000;

Boyer, 2001). This research is the first to offer a psychological explanation for the trends in

reincarnation practices documented around the world and especially in Eastern traditions. More

broadly, the research reported here serves as a case study for why scholars in the scientific study of

religion need to meaningfully engage with research on how humans think if they are to explain the

recurrence of religious ideas across cultures (see, for example, McCauley, 2012; Slingerland,

2012). Practices designed to establish who has returned to the human world are underpinned by

similar psychological assumptions about what constitutes evidence of this fact. These processes

do not concern the trivial details of parochial traditions; rather, they determine the features of

social systems of tribal and world religions: the distribution of names, titles, property inheritance,

wealth, and even political governance. If we want to explain religion, we need to understand the

influence of cognition as well as culture.

Acknowledgements

Thanks to Jesse Bering and Paulo Sousa for their guidance in the early stages of this project;

Shaun Nichols for reading and commenting upon an earlier version of this manuscript; the two

anonymous reviewers and editor Richard Sosis for their thorough and constructive feedback; the

students at the Cognitive Science of Religion Lab Group at California State University,

Northridge, for their assistance with the preparation of this manuscript; and to the people of

Chennai for their participation and hospitality. This research was funded by a Department for

Employment and Learning Research Studentship (DEL) award at Queen’s University, Belfast and

written during release time financially supported by California State University, Northridge.

References

Addis, R. D., Tippett, L. J. (2004). Memory of myself: Autobiographical memory and identity in

Alzheimer's disease. Memory, 12, 56-74.

Andersen, S. M. & Cole, S. W. (1990). "Do I know you?" The role of significant others in general

social perception. Journal of Personality and Social Psychology, 59, 384-399.

Andersen, S. M., Glassman, N. S. & Chen, S. (1995). Transference in social perception: the role of

chronic accessibility in significant-other representations. Journal of Personality and Social

Psychology, 59, 384-399.

Astitu, R., Harris, P. L. (2008). Understanding mortality and the life of the ancestors in rural

Madagascar. Cognitive Science, 32, 713-740.

Barrett, J. L. (2000). Exploring the natural foundations of religion. Trends in Cognitive Sciences,

4(1), 29-34.

Barrett, J. L. (1999). Theological correctness: Cognitive constraint and the study of religion.

Method and Theory in the Study of Religion, 37, 608-619.

Barrett, J. L. (1998). Cognitive constraints on Hindu concepts of the divine. Journal for the

Scientific Study of Religion, 37(4), 608-619.

Barrett, J. L., Keil, F. C. (1996). Anthropomorphism and God concepts: Conceptualizing a non-

natural entity. Cognitive Psychology, 31, 219-247.

Bastian, M. (2002). Irregular visitors: Narratives about ogbaanje (spirit children) in Southern

Nigerian popular writing. In Newell (Eds.) Readings in African popular fiction. Indiana

University Press.

Bell, C. (1931). The religion of Tibet. London: The Clarendon Press.

Bering, J. M. (2002). Intuitive conceptions of dead agent's minds: The natural foundations of

afterlife beliefs as a phenomenological boundary. Journal of Cognition and Culture, 2, 262-

Bering, J.M., Bjorklund, D.F. (2004). The natural emergence of reasoning about the afterlife as a

developmental regularity. Developmental Psychology, 40, 217-233.

Blok, S., Newman, G. & Rips, L. J. (2005). Individuals and their concepts. In W-k, Ahn, R.L.,

Goldstone, B.C., Love, A. B. Markman, & P. Wolff (Eds.). Categorization inside and outside

the lab. Washington DC: American Psychological Association

Bloom, P. (2004). Descartes' baby: How child development explains what makes us human.

London: William Heinemann Ltd.

Bose, A., Singh, V. K., & Adhikary, M. (1991). Demographic diversity of India 1991 census: state

and district level data. A reference book.

Boyer, P. (2003). Are ghost concepts “intuitive,” “endemic” and “innate”. Journal of Cognition

and Culture, 3, 233-243.

Boyer, P. (2001). Religion explained. New York, NY: Basic Books.

Butler, J. (1819). Of personal identity. In J. Angus (Ed.), The analogy of religion (pp. 211–215).

London, UK: Allman and Sawers.

Cohen, E. (2007). The mind possessed. New York: Oxford University Press.

Cohen, E. & Barrett, J. (2011). Folk anthropology: The cognitive anthropology of the person. In J.

Wentzel van Huysteen & Erik Wiebe (Eds.) In search of self: Interdisciplinary perspectives on

personhood. Grand Rapids, CA: Wm. B. Eerdmans Publishing Co.

Cohen, E., & Barrett, J. (2008). When minds migrate: Conceptualizing spirit possession. Journal

of Cognition and Culture, 8, 23-48.

Cohen, E., Burdett, E., Knight, N., & Barrett, J. (2011). Cross‐Cultural similarities and differences

in person‐body reasoning: Experimental evidence from the United Kingdom and Brazilian

Amazon. Cognitive Science, 35(7), 1282-1304.

Conway, M. A. (2005). Memory and the self. Journal of Memory and Language, 53(4), 594-628.

Creider, J. T. Two lives: My spirit and I. (1986). Massachusetts: Interlink Publishing.

Dundas, P. (2002). The Jains. New York: Routledge.

Fiske, S. T., Taylor, S. E. (1991). Social cognition. New York: McGraw-Hill.

Frazer, J. G. (1918). Folk-Lore in the Old Testament: Studies in comparative religion, legend and

law (Vol. 3). Macmillan.

Gelman, S. (2005). The essential child: Origins of essentialism in everyday thought. Oxford

University Press.

Gobbini, A. I., Leibenluft, E., Santiago, N. & Haxby, J. V. (2004). Social and emotional

attachment in the neural representation of faces. Neuroimage, 22, 1628-1635.

Goody, J. (1962). Death, Property and the Ancestors: A Study of the Mortuary Customs of The

Lodagoa of West Africa. Stanford University Press.

Gottlieb, A. (2004). The afterlife is where we come from. Chicago, Ill.: University of Chicago

Press.

Gupta, A. (2002). Reliving childhood? The temporality of childhood and narratives of

reincarnation. Ethnos, 67, 33-56.

Gutheil, G. Rosengren, K.S. (1996). A rose by any other name: preschoolers' understanding of

individual identity across name and appearance changes. British Journal of Developmental

Psychology, 14, 477-498.

Guthrie, S. E. (1993). Faces in the clouds: A new theory of religion. New York: Oxford

University Press.

Haraldsson, E., & Samararatne, G. (1999). Children who speak of memories of a previous life as a

Buddhist monk: Three new cases. Journal of the Society for Psychical Research, 63(857), 268-

Henrich, J., Heine, S. J., & Norenzayan, A. (2010). Most people are not WEIRD. Nature,

466(7302), 29-29.

Kaplan, F. E. S. (2000). Some thoughts on ideology, beliefs, and sacred kinship among the Edo

(Benin) people of Nigeria. In Brown, C.K., & A.P. Thakur (eds.). Kinship, marriage and the

family. London: Oxford University Press.

Kaufman, F. (2004). Pre-vital and post-mortem non-existence. In D. Benatar (Ed.), Life, death, &

meaning: Key philosophical readings on the big questions (pp. 241-264). Oxford, UK:

Rowman & Littlefield Publishers, Inc.

Klein, S. B., Cosmides, L., Tooby, J., & Chance, S. (2002). Decisions and the evolution of

memory: Multiple systems, multiple functions. Psychological Review, 109(2), 306-329.

Klein, S. B., & Nichols, S. (2012). Memory and the sense of personal identity. Mind, 121(483),

677-702.

Leveroni, C. L., Seidenberg, M., Mayer, M. R., Mead, L. A., Binder, J. B. & Rao, S. M. (2000).

Neural systems underlying the recognition of familiar and newly learned faces. The

Journal of Neuroscience, 20(2), 878-886.

Locke, J. (1975). An essay concerning human understanding. Oxford, UK: Clarendon Press.

(Original work published 1690)

Macdonald, C. (2005). Varieties of things: Foundations of contemporary metaphysics. Oxford:

Basil Blackwell.

Malle, B. F. (2004). How the mind explains behavior: Folk explanations, meaning, and social

interaction. MIT Press.

Malle, B. F., Moses, L. J., & Baldwin, D. A. (Eds.). (2001). Intentions and intentionality:

Foundations of social cognition. MIT press.

Matlock, J. (1993). A Cross-cultural study of reincarnation ideologies and their social correlates

(Unpubplished master’s thesis). Hunter College, New York.

Matlock, J. G., & Mills, A. (1994). A trait index to North American Indian and Inuit reincarnation

beliefs. In A. Mills & R. Sobodin (Eds.), Amerindian rebirth: Reincarnation belief among

North American Indians and Inuit (pp. 299-356). Toronto, Canada: University of Toronto

Press.

McCauley, R. N. (2012). A cognitive science of religion will Be difficult, expensive, complicated,

radically counter-intuitive, and possible: A response to Martin and Wiebe. Journal of the

American Academy of Religion, 80(3), 605-610.

Meltzoff, A. N., & Moore, M. K. (1995). Infants' understanding of people and things: From body

imitation to folk psychology. In J.L. Bermudez, A. Marcel, and N. Eilan (Eds.), The Body and

Self (pp. 43-69), MIT Press: Bradford.

Mills, A., & Slobodin, R. (Eds.). (1994). Amerindian rebirth: Reincarnation belief among North

American Indians and Inuit. Toronto, Canada: University of Toronto Press.

Nagel, T. (1986). The view from nowhere. Oxford, UK: Oxford University Press.

Nichols, S. (2007). Imagination and Immortality: Thinking of Me. Synthese, 159(2), 215-233.

Nicholson, H. (2014). Social Identity Processes in the Development of Maximally Counterintuitive

Theological Concepts: Consubstantiality and No-Self. Journal of the American Academy of

Religion, 82(3), 736-770.

Obeyesekere, G. (2002). Imagining karma: Ethical transformation in Amerindian, Buddhist, and

Greek rebirth. London: University of California Press.

Obeyesekere. (1994). Foreword: Reincarnation eschatologies and the comparative study of

religions. In A. Mills & R. Sobodin (Eds.), Amerindian rebirth: Reincarnation belief among

North American Indians and Inuit (pp. xi - xxiv). Toronto, Canada: University of Toronto

Press.

Oluwole, S. B. (1992). Reincarnation: An issue in African philosophy? In Oluwole, S.B. (1992).

Witchcraft, reincarnation, and the god head. Lagos: Excel Publishers.

Orona, C. J. (1990). Temporality and identity loss due to Alzheimer's disease. Social Science and

Medicine, 30, 1247-1256.

Orubu, A. O. (2001). The hereafter in African traditional religions. in A. O. Orubu (ed.), African

Traditional Religion. Benin City: Institute of Education, University of Benin.

Parfit, D. (1984). Reasons and persons. New York, NY: Oxford University Press.

Parkin, R. (1988). Reincarnation and alternate generation equivalence in Middle India. Journal of

Anthropological Research, 1-20.

Penelhum, T. (1970). Survival and disembodied existence. London, UK: Routledge.

Perry, J. (1978). A dialogue on personal identity and immortality. Indianapolis, IN: Hackett

Publishing Company, Inc.

Perry, J. (1975). Personal identity. University of California Press.

Pyysiäinen, I. (2004). Intuitive and explicit in religious thought. Journal of Cognition and Culture,

4, 123-150.

Reid, T. (1969). Essays on the intellectual powers of man. Cambridge, MA: MIT Press. (Original

work published 1785)

Ronch, J. L. (1996). Mourning and grief in late life Alzheimer's dementia: Revisiting the vanishing

self. American Journal of Alzheimer’s Disease and Other Dementias, 11, 25-28.

Rosengren, K. S., Gelman, S. A., Kalish, C. W. and McCormick, M. (1991). As time goes by:

Children’s early understanding of growth in animals. Child Development, 62, 1302-1320.

Shah, N. J., Marshall, J. C., Zafiris, O., Schwab, A., Ziles, K. & Fink, G. R. (2001). The neural

correlates of person familiarity: A functional magnetic resonance imaging study with clinical

implications. Brain, 124, 804-815.

Shoemaker, S. (1963). Self-knowledge and self-identity. Ithaca, NY: Cornel University Press.

Shoemaker, S., & Swinburne, R. (1984). Personal identity. Oxford, UK: Blackwell.

Slingerland, T. (2012). Back to the Future: A Response to Martin and Wiebe. Journal of the

American Academy of Religion, 80(3), 611-617.

Slone, J. (2007). Theological Incorrectness: Why Religious People Believe What They Shouldn't.

Oxford University Press.

Somersan, S. (1981). Death symbolism: A cross-cultural Study. Cross-Cultural Research, 29(2),

158-177.

Sperber, D. (1996). Explaining culture: A naturalistic approach. Oxford, UK: Blackwell

Publishing.

Spiro, M. E. (1982). Buddhism and society: A great tradition and its Burmese vicissitudes. Univ of

California Press.

Swanson, G. (1960). The birth of the Gods: The origin of primitive beliefs. Ann Arbor: University of

Michigan Press.

Tenzin, G. (1990). Freedom in Exile: The Autobiography of the Dalai Lama. New York:

HarperCollins.

Tulving, E. (1972). Episodic and semantic memory. In E. Tulving & W. Donaldson (Eds.),

Organization of memory. New York: Academic Press, 1972.

White, C. (in press). Establishing personal identity in reincarnation: Minds and bodies

reconsidered. Journal of Cognition and Culture.

White, C. (2014). The cognitive foundations of reincarnation: ethnographic and experimental

evidence. Manuscript submitted for publication.

White, C., Sousa, P., & Berniunas, R. (2014). Psychological essentialism in selecting the 14th

Dalai Lama: An alternative account. Journal of Cognition and Culture, 14, 1-2, 157-158.

White, C., Kelly, B., & Nichols, S. (In press). Remembering past lives: Why memories convince

people that they have lived before. In Cruz, H. & Nichols, R. (Eds.), The Cognitive Science of

Religion and its Philosophical Implications. London: Bloomsbury Publishing.

Williams, B. (1973). Problems of the self: Philosophical papers 1956-1972. Cambridge, UK:

Cambridge University Press.

White, C. (In press). Cross-cultural Similarities in Reasoning about Personal Continuity in...

Documents

Transcript of White, C. (In press). Cross-cultural Similarities in Reasoning about Personal Continuity in...

Continuity In Leibniz's Mature Metaphysics

Elliptic Systems, Singular Sets and Dini Continuity

MaineDOT COVID-19 Continuity of Operations - sicop

Taking Reincarnation Seriously: Critical Discussion of Some Central Ideas from John Hick

The Architecture of Continuity: Essays and Conversations

Digital Wallpaper: Tiles of Proliferation and Continuity

Draft Continuity Plan - LawPhil

Therapists and their patients: Similarities and differences in ...

Similarities between Psychology and Classical Taoism

myth, metatext, continuity and - CiteSeerX

Achieving Care Continuity - Advisory.com

Question format and the distribution of belief in reincarnation ...

5G: Service Continuity in Heterogeneous Environments

Similarities of Retinoblastoma Function between Plants and ...

Special Capital Expenditure Budget: Business Continuity Facility

Single Radio Voice Call Continuity - Cisco

Chapter 1 Limits and Continuity

Business Continuity and Emergency Preparedness Planning

INTERTEXTUAL CONTINUITY IN SUPERGIRL STORIES ...

Limits, Continuity and