MEMORY IN TOUCH - Psicothema

Does the modality in which informationis presented influence how it is remembe-red? Human information tends to be organi-sed predominantly in either verbal or spatial

Psicothema, 1999. Vol. 11, nº 4, pp. 747-767ISSN 0214 - 9915 CODEN PSOTEGCopyright © 1999 Psicothema

Psicothema, 1999 747

MEMORY IN TOUCHSusanna MillarOxford University

Models of short-term memory have to take into account that touch is not a tightly organised sin-gle modality. Touch, without vision or other external cues, depends on information from touch, move-ment and from body-centred (posture) cues. These inputs vary with the size and types of object, and withtask demands. It is argued that the convergence and overlap of inputs from different sources is crucialto parsimonious organisation for memory and recall. «Modality-specific» input conditions thus form anintegral part of the available information, which changes, and is changed by longer-term information.Three general principles apply: (i) Parsimony of coding modality-specific inputs for recognition and re-call; (ii) links with output systems which can «rehearse» information; and (iii) longer-term familiaritywith procedures and types of coding. The introduction and first section discuss these points in relationto models for hearing and vision. The third section cites findings on modality-specific tactual memory,and explains tactual memory spans in terms of paucity versus redundancy of reference information toorganise inputs spatially. Movements are considered next as inputs and as spatially organised outputsthat can provide haptic rehearsal. The final section argues that intersensory modality-specific processesand longer-term memory need to be included as interrelated systems in STM models in order to accountfor memory in touch.

La memoria en el tacto. Los modelos de memoria a corto plazo deben tener en cuenta que el tac-to no es una modalidad única estrechamente organizada. El tacto sin visión u otras claves externas, de-pende de la información obtenida a partir del tacto, del movimiento y de las claves (la postura) centra-das en el cuerpo. Estos inputs varían con el tamaño y los tipos de objetos, y con las demandas de la ta-rea. La convergencia y el solapamiento de los inputs a partir de diferentes fuentes es crucial para la or-ganización de la memoria y el recuerdo. Las condiciones del input «específicas de la modalidad formanuna parte integral de la información disponible, que cambia, y es cambiada por la información a largoplazo. Tres principios generales pueden aplicarse: (i) La parsimonia de la codificación de los inputs es-pecíficos de la modalidad para el reconocimiento y el recuerdo; (ii) las uniones con los sistemas del out-put que pueden «repetir» la información; y (iii) mayor familiaridad con los procedimientos y los tiposde codificación. En la introducción y en la primera sección se discuten estos puntos en relación con losmodelos de visión y audición. La tercera sección cita resultados sobre la memoria del tacto específicade la modalidad, y explica la amplitud de la memoria táctil en función de la pobreza versus la redun-dancia de información de referencia para organizar los inputs de manera espacial. Después se conside-ran los movimientos como inputs y outputs organizados espacialmente que pueden proporcionar repeti-ción háptica. En la sección final se argumenta que los procesos intersensoriales específicos de la moda-lidad y la memoria a largo plazo necesitan ser incluidos como sistemas interrelacionados en los mode-los de MCP para poder explicar la memoria en el tacto.

Correspondencia: Susanna MillarDepartment of Experimental PsichologyOxford UniversitySouth Parks Road OX1 3UDOxford (United Kingdom)

form. Interestingly, each is connected with aparticular sense modality: Space with vi-sion, language with hearing. Touch does nothave such a well-defined link with a parti-cular form of knowledge. Feeling texture,temperature, pressure and pain in whichtouch is specialised, belong more to the«qualia» of experience —as do colour in vi-sion, or pitch in hearing—, than to specifi-cally linguistic or spatial forms of organi-sing information. The links between visionand space, hearing and language are not, ofcourse, exclusive. Visual icons, scripts,signs and gestures can carry verbal informa-tion. Distance, direction, and even shapecan be remembered from hearing sound pat-terns and verbal descriptions. Touch canconvey both forms of information.

Is there, therefore, any need to includethe perceptual modality of inputs in modelsof memory? I shall argue that there is. The«medium» is not the «message». But theperceptual conditions in which inputs occur,and the links between perception and outputsystems, are an integral part of the type andamount of information that is available, andhow that information is processed and re-membered.

The general framework I prefer for des-cribing information processing and memoryuses the metaphor of dynamic networks ofinterrelated and converging processes (Mi-llar, 1994, 1997). It focuses on active pro-cessing of the information that is availableto the organism from external and internalsources. The contrast is with metaphors thatimply a static architecture. Such metaphorsunderplay the continual, reciprocal influen-ce of previous and current information inprocessing, though that is obvious even inorder effects in experimental presentations.Architectural metaphors suggest a rigid hie-rarchical («bottom-up» or «top-down») or-ganisation. Separate inputs from the moda-lities are either integrated at the top level bya separate translation mechanism, or simply

subserve a top-level abstract descriptionthat obtains for all inputs. Metaphors thatimply active networks of interacting andconverging processes accord better with cu-rrent findings on neural connections in theconstantly active brain. Thus inputs frommultisensory sources seem to converge invarying combinations in a number of brainareas. Above all, the metaphor of active,converging processes is needed for touch.

Touch is not a single modality, nor evena single perceptual system. What we call«touch» refers to combinations of a numberof different, converging sources of informa-tion that include inputs from some skin re-ceptors to greater or lesser extents. The no-tion of perceptual systems that receive in-puts from several sources is true also forother perceptual systems (Gibson, 1966).But in touch, the combinations of inputsvary with the type, size, meaning and fami-liarity of objects, as well as with current in-formation, and importantly also, with taskdemands. The questions we ask about me-mory in touch thus have to be seen in thecontext of the combinations of disparate in-puts that need to converge for recognitionand short-term recall in different haptic con-ditions.

Because differences in informationalconditions are particularly important intouch, the present paper focuses on short-term memory for raised line configurations,rather than on object recognition. The neu-ral pathways in identifying objects (what?)and locations (where?) are not precisely thesame (e.g. Anderson, 1987; Schneider,1967; Weiskrantz, 1986). Recognising fa-miliar manipulable objects, that have well-known names, taps into a wide range of se-mantic knowledge and meaningful verbaldescriptions of their every day use and eventheir visual appearance for subjects who ha-ve the experience. Prior knowledge has tobe considered in examining short-term me-mory. But it is more circumscribed, and so-

MEMORY IN TOUCH

748 Psicothema, 1999

mewhat differently focused, in short-termmemory tasks that demand recognition orrecall of relatively unfamiliar tactual raisedline and dot patterns. The present paper cen-tres on the information that is available tosubjects in such haptic conditions.

The paper is divided into several sec-tions. The first section briefly describes so-me relevant theoretical models of short-term memory. Since they are predominantlybased on findings with visual and verbalmaterials, these are considered first to esta-blish the context in which findings on me-mory for touch must be examined. The nexttwo sections survey findings from studieson memory for what is generally called«touch» although it concerns informationfrom a number of different sources that in-clude touch. The final section considers thefindings in relation to intersensory informa-tion, and looks at implications of the fin-dings for theoretical descriptions.

Some Theoretical Considerations

Models of memory have often relegatedthe modality of inputs to a very minor role.Some equate modality-specific informationsolely with brief persistence of stimulationafter its offset, which has no function in me-mory except that of initial registration. Pro-cessing is carried by a short-term memorysystem that is strictly limited in attentionalcapacity. The number of serial items towhich it can attend depends crucially on re-coding the inputs into more economicalforms (Miller, 1956). A favoured descrip-tion for many years was of fast-fading tra-ces of the stimulation that must be transla-ted quickly into verbal form, in order to bemaintained in limited-capacity short-termmemory, prior to being stored in longer-term memory (e.g. Atkinson & Shiffrin,1968).

The most influential recent model is a«working memory» system which has seve-

ral components (Baddeley, 1990). It assu-mes a central executive, decision-makingprocess that allocates attentional resources.A speech-based articulatory loop refreshesfast-fading phonological (heard) inputs th-rough rehearsal (Baddeley, 1986, 1990;Baddeley & Hitch, 1974). A «visuo-scratch-pad» accounts for findings which show thatmemory for spatial location, direction andshape is disrupted more by conflicting vi-suo-spatial than by verbal information, andfor evidence that such non-verbal informa-tion can be maintained actively in memoryover short delays (e.g. Atwood, 1971; Bro-oks, 1968; Hitch & Halliday, 1983; Hitch etal, 1988; Kosslyn, 1980; Kosslyn et al,1978; Millar, 1972a; Segal & Fusella, 1970;Shepard & Cooper, 1982; Shepard & Feng,1972)).

It is noteworthy that the main burden ofmaintaining phonological information intemporary memory is assigned to speech-based output activities, rather than to inputcoding (e.g. Baddeley, 1986; 1990; Broad-bent, 1984; Conrad, 1964; Monsell, 1987).There is ample evidence that concomitantspeech, including mouthing of an irrelevantsyllable, disrupts recall for heard speech. Ithas been much more difficult to demonstra-te temporary memory for modality-specificinput sounds. Better recall of the last item ina heard than viewed series (e.g. Waugh &Norman, 1965) is sometimes attributed tothe persistence of the last sound in uncoded(«raw») «echoic» form. The notion of pre-categorical acoustic storage subsumes suf-fix effects in which an irrelevant speechsound at the end of an auditory sequence af-fects recall of the whole series. However,phonological effects in memory have alsobeen found when speech output is made dif-ficult, or is impossible (e.g. Besner, 1987;Vallar & Baddeley, 1982). Articulation rateis related to the size of immediate memoryspans (e.g. Baddeley, 1990; Hitch & Halli-day, 1983; Hitch et al, 1988, Hulme & Tor-

SUSANNA MILLAR


doff, 1989). But age differences in span arenot eliminated by equating articulation ra-tes, as would be expected if speech outputwere the only factor (Henry & Millar, 1991,1993). Familiarity with input sounds hassignificant effects on the size of immediatememory spans with age (Henry & Millar,1991). Very young children tend not to useverbal rehearsal strategies (e.g. Conrad,1971), but their memory span for sounds in-creases nevertheless to some extent. Thecontribution of longer-term memory to in-put coding for temporary memory is, there-fore, a further factor that has to be includedin theoretical models (Henry & Millar,1993). Familiarity with inputs, as well aslinks with output systems thus need to beconsidered in examining findings on touch.

Experimental evidence for temporarymemory for visual inputs has been even mo-re difficult to establish than for sounds. Vi-sion does not seem to have such an obviousclosely-knit output system as vocalisationfor hearing, though gestures in pointing, re-aching, drawing and locomotion constitutesuch output systems. However, the «percep-tion-action» models, deriving from Gibson(1979), which stress input-output links, usethese to explain perceptual organisation wit-hout recourse to memory processes. In stu-dies of visual memory, on the other hand,the main controversy centred on the ques-tion whether visual memory is visual, orspatial and «abstract» in character. The vi-sual character of introspectively vivid vi-suo-spatial imagery seems obvious to thosewho experience it. Moreover, errors in «re-ading off» from very vivid or «eidetic» ima-gery show that it involves memory ratherthan persistence of «raw» sensory stimu-lation (Haber & Haber, 1964). Experimentalfindings have also been interpreted in ana-logy with a transitory quasi-«pictorial» vi-sual buffer that functions as if in «a me-dium» of coordinated space (e.g. Kosslyn,1980, 1981). But experiments have not al-

ways distinguished between visual and spa-tial aspects of the materials. Discrepant vi-sual inputs that interfere with memory forvisual presentations may do so because oftheir spatial form, and not because they arevisual in character (e.g. Logie, 1986; Logie,Zucco & Baddeley, 1990). Thus, evidencethat simple flashes of light have relativelylittle effect on memory for visuo-spatial in-formation has suggested that memory de-pends on «abstract» spatial representations.The notion implies that representations inmemory are either not specifically visual incharacter (introspective reports are mista-ken), or that their visual character is func-tionally irrelevant, or both.

Nevertheless, the total burden of currentevidence suggests that temporary memoryfor modality-specific visual aspects, as wellas for their spatial organisation, may be as-sumed. The fact that electrical stimulationof visual areas of the brain produces visualimages (e.g. Dobelle, Mladejovsky & Gir-vin, 1974), and cognate evidence for the in-volvement of the visual cortex in attemptsto form visual images (Goldenberg et al,1990), suggests a physiological basis for«visual» aspects of introspectively reportedimagery. Whether or not that is sufficient toestablish that the visual character of repor-ted visual imagery is biologically useful, isa moot point. But further experimental stu-dies with children and adults also strengthenevidence for temporary memory for specifi-cally visual aspects, as well as for spatial in-formation (e.g. Logie & Baddeley, 1990;Logie & Pearson, 1997). Thus, recognitionas well as recall for visual patterns was bet-ter than memory for sequences of move-ments to spatial targets, and the differencevaried with age (Logie & Pearson, 1997).Taken together, the findings suggest a so-mewhat more complex picture of «workingmemory» than the original notion of short-term memory as a single system with one li-mited attention bottleneck.

MEMORY IN TOUCH


An alternative has been to assume com-pletely distributed storage systems with mul-tiple, separate «processing modules», eachwith its own temporary storage capacity(e.g. Monsell, 1984). In principle, modality-specific aspects of memory could easily beaccommodated in such a system, though thatwas not the reason for the notion. The extre-me version assumes quite separate «modu-les» for different cognitive processes (Fodor,1983). It was originally taken up with greatalacrity, because the considerable functionalspecialisation of different areas of the brainseemed to accord better with descriptions interms of separate «modules» than with theidea of one over-arching cognitive system.The evidence for regional cerebral speciali-sation of functions is not in doubt. However,the physiological and psychophysiologicalevidence neither requires, nor is actuallycompatible with the assumption that thereare only very limited interactions betweendifferent brain regions (e.g. Squire, 1987).

It can be argued, on the contrary, that thefindings on specialisation of functions arebetter described in terms of the convergen-ce, in different combinations for differentregions, of processing paths or activities ari-sing from specialised analyses. A more dy-namic picture of this kind seems to be re-quired particularly to account for evidenceon memory processes in touch.

Touch has been comparatively neglectedin studies of psychology. The issues that ha-ve been considered here so far were raisedpredominantly by findings with visual andverbal materials. How do findings on touchfit in with these? Useful general theoreticalmodels should account for findings ontouch as well as on vision and hearing. Butit is not sufficient to give theoretical des-criptions only at the most general level. Thequestion is what forms of information canbe assumed to be available for processing intactual conditions, and what, if any, effectsthese have on memory.

The point is that «touch» is actually aeuphemism for intersensory processing ofinformation from a number of different sour-ces. The skin receptors convey informationabout texture, pressure, temperature andpain, as well as light touch, and the inputsoccur in various combinations (e.g. Katz,1925). Moreover, even passive touch can de-pend on concomitant input from propriocep-tive sources. Single unit recording hasshown that discrimination of round and rec-tangular objects held in the monkey’s handdepends on converging inputs from touch re-ceptors in the palm and proprioceptive in-puts from the posture of the finger joints (Sa-kata & Iwamura, 1978). The organismquickly adapts to passive touch or pressure.Even passive perception usually needs inter-mittent, necessarily sequential, stimulation.Movement is crucial for active tactual ex-ploration to gain information about objectsand environments (e.g. Gibson, 1962; Re-vesz, 1950). The term «haptics» was used todesignate that proprioceptive and kinaesthe-tic inputs function together with touch in ex-ploring objects (Revesz, 1950). Gibson(1966) made the important point that themodalities of vision and hearing, as well astouch, are perceptual systems which dependon several sources of information.

However, in touch or haptics the combi-nation of inputs from different sources doesnot seem to operate as an inherently tightlyorganised single perceptual system. Explo-ratory skill, experience with the stimulus ob-jects and with tasks, as well as physical cha-racteristics of stimulus objects determinehaptic movements and performance (e.g.Appelle, 1991; Berla & Butterfield, 1977;Davidson, 1972). The combinations of con-verging information, from touch, finger,hand and arm movements, but also proprio-ceptive inputs from hand and/or body postu-res, differ not only with task demands, butalso with the size, depth and composition ofobjects in touch (Millar, 1981b, 1994, 1997).

SUSANNA MILLAR


Evidence on whether, and if so how, thevarious modality-specific aspects of haptictask conditions affect memory is consideredin the next section in relation to temporaryor «working» memory.

Short-term tactual memory

Evidence for modality-specific tactileshort-term memory was found in a studywhich demonstrated a tactile suffix effect(Watkins & Watkins, 1974). Recall of theserial position in which fingers of bothhands had been touched was disrupted by astroke across two fingers at the end of a se-ries. The finding was interpreted as eviden-ce for capacity-limited tactile memory, ana-logous to suffix effects in auditory memory(see earlier). But the finding raises similarquestions as demonstrations for visuo-spa-tial inputs. Thus, it is possible to argue thatrecall of the sequence in which fingers hadbeen touched depended on coding the spa-tial location of the fingers that had been tou-ched. In principle, finger locations can becoded spatially by reference to body-cen-tred frames, in the absence of vision, andeven if covert visuo-spatial strategies werenot being used. Memory may, therefore, ha-ve depended on the spatial organisation ofthe tactual inputs, rather than on memoryfor «feels».

Millar (1975 a) showed that modality-specific tactual effects could be separatedexperimentally from verbal recoding of theinputs. Recall spans were compared for listsof braille patterns which were either similartactually, or had similar sounding names, orwere dissimilar on both counts. Subjectswere congenitally totally blind children atdifferent stages of learning braille. They feltthe patterns sequentially in each list. Thetask was to point to the serial position that asubsequently presented test pattern had oc-cupied in the memory series. Results for thetactually similar lists were completely the

opposite of those for phonologically similarlists. As expected from Conrad’s (e.g. 1964,1971) work with nameable visual items,Millar (1975 a) found that recall spans forpatterns with phonologically confusable na-mes were significantly impaired comparedto patterns with dissimilar names, showingthat the items had been recoded verbally.The difference related significantly, notonly to the size of memory spans on controllists, but also to pre-test naming speeds forthe items. Subjects with faster pre-test na-ming latencies showed larger recall spansand were more impaired by phonological si-milarity. By contrast, tactual similarity ef-fects were associated with small recallspans, and related inversely to pre-test na-ming. Tactual similarity impaired recall mo-re by subjects with smaller recall spans whoshowed little or no phonological similarityeffects, and were either slow on pre-test na-ming or could not name the items on pre-test.

Evidence that tactual similarity affectsthe size of recall spans suggests temporarymemory for tactual aspects of the items. Thestudy has been described in some detailagain, because it is possible to rule out ex-planations in terms of visual, as well as pho-nological strategies, since the subjects werecongenitally totally blind. But that does notmean that tactual memory effects are confi-ned to blind children. Blindfolded sightedyoung children were tested on recall spansfor lists of tactually presented objects. Theobject series were again either similar in fe-el or had similar names (Millar, 1975 b). Asfor the blind, tactual similarity, but not pho-nological similarity reduced the recall spansof children with small spans in control con-ditions, while subjects who achieved largespans in control conditions were signifi-cantly affected by phonological and not bytactual similarity in the memory lists. Seve-ral subsequent studies also showed that thefindings could not be attributed to indivi-

MEMORY IN TOUCH


dual difference factors, such as age (Millar;1978 a,b). Thus, the same children who hadrecall spans of six or more tactual items thatthey could name, and improved furtherwhen the items were grouped, produced re-call spans of only two to three items for tac-tual nonsense shapes. If anything, groupingtactual nonsense shapes had the opposite ef-fect. The findings thus suggest that moda-lity-specific tactual memory must be assu-med, but it typically consists of only two tothree tactual items.

Prima facie, there seems to be no reasonwhy memory spans for tactual patternsshould be worse than for the same patternsin vision, if the tactual patterns are codedspatially as global shapes. Evidence for spa-tial coding comes from studies of brain in-sults in specialised cerebral regions that areknown to be important in visuo-spatial or-ganisation (e.g. right hemisphere, post-pa-rietal area). Such infarcts also disrupt tem-porary memory for nonverbal tactual inputsin spatial tasks, although it is not alwaysclear whether the loss is primarily sensoryor attentional (e.g. Stein, 1991). But there isgood evidence that spatial tasks involve anumber of cortical and subcortical regionsof the brain. The involvement of differentregions (e.g. hippocampal, sensori-motor,somatosensory, pre-frontal, post-parietal,cerebellar, occipital and temporal) seems todepend on the combination of demands thattasks make on spatial, verbal and cognitiveand memory skills, and on biomechanicalconstraints that particular conditions invol-ve. Spatial tasks generally show greater in-volvement of the right hemisphere, whetherthe stimulus materials are visual or tactile.However, it is not possible to make the re-verse inference, and to conclude that a left-hand advantage necessarily involves righthemisphere/spatial processing. Similarly,without additional evidence and adequatetask analysis, a right hand advantage doesnot necessarily imply left hemisphere/ver-

bal processing. Thus, braille studies havereported left hand advantages, right handadvantages, and equal performance by bothhands. Not surprisingly, the studies differwidely in the type of braille task that wasbeing used, and in the experience and skillof the subjects who took part. Moreover, so-me apparently spatial tasks, such as aiming,have produced right hand advantages (Mi-llar, 1994, for review). To make sense, sti-mulus conditions, task demands, and sub-jects’ experience with exploratory strategiesand materials have to be taken into account(e.g. Wilkinson & Carr, 1987).

There has long been evidence that me-mory for unfamiliar tactual shapes is muchless efficient than memory for the same un-familiar shapes in vision (e.g.; Berla & But-terfield, 1977; Gilson & Baddeley, 1969;Goodnow, 1971; Millar 1971a, 1977a,1978b, 1990, 1991). Such assertions are of-ten misunderstood. It should be stressed atonce, therefore, that they do not, of course,call in question that shape «can» be percei-ved by touch. That has long been establis-hed (e.g. Gibson, 1962; Katz, 1925), and re-quires no further research. Nor should evi-dence of poor tactual recognition be takento mean that memory must necessarily bepoor when the inputs are tactual. For instan-ce, tactual recognition of familiar three-di-mensional objects is often very good (e.g.Hatwell, 1978; Katz, 1925; Klatzky, Leder-man & Metzger, 1985; Weber, 1834, transl.1978). Levels of efficiency, whether high orlow, are of theoretical interest only becausethey raise questions about the factors thatproduce differences. The question here isthus about the informational conditions thatunderlie relatively small spans and poor me-mory for tactual inputs.

Thus, it is relevant that poorer tactual re-cognition is more often found for unfamiliartwo-dimensional raised line and raised-dotpatterns, early in learning (e.g.; Goodnow,1971; Hatwell, 1978; Lederman, Klatzky,

SUSANNA MILLAR


Chataway & Summers, 1990; Millar,1975a,b, 1977b, 1978a; 1985a, 1990, 1991),rather than for familiar objects (see earlier),and for three-dimensional shapes (David-son, Barnes & Mullen, 1974; Klatzky et al,1985; Millar, 1974; Shimizu, Saida, & Shi-mura, 1993).

The difference is not merely that solidforms, as such, necessarily provide moregeneral information than outline shapes, ashas sometimes been assumed. Many studiesof 3-D shapes use familiar objects. For therecognition of such objects, shape (e .g.long versus round) is merely one amongmany possible cues, from temperature to re-silience to pressure, to the use, name andcontextual meaning of the object (see ear-lier). The important advantage for recogni-sing the shape of unfamiliar 3-D versus 2-Dforms is that 3-D shapes potentially affordmore reference cues for spatial coding. Thusboth hands are usually used to manipulatesmall three-dimensional forms. The twohands can act as spatial anchors and re-ference frames in relation to each other tolocate component features. For large, statio-nary three-dimensional objects that are ex-plored by hand and arm movements, con-tour features can be located and spatially or-ganised by reference to body-centred re-ference frames. Two-dimensional raised li-ne configurations are usually explored byone finger, and, in particular if they aresmall, are difficult to relate to body-centredframes.

We actually have sufficient evidencefrom past findings to be able to specify themain types of information that can vary intasks that demand tactual memory for shapeor form. Previous experience, both with andwithout vision, affects performance signifi-cantly. Modes of exploration and move-ments vary with familiarity, and with the ty-pe, size, depth and composition of stimuli.These, and task conditions are among themost important variables in haptic tasks

(Davidson, 1972, 1974; Davidson et al,1974; Locher & Simmons, 1978; Millar,1981, 1988a; Simmons & Locher, 1979). Acrucial factor is the type, amount and con-vergence of reference information that isavailable from internal and external sour-ces, since it determines the spatial organisa-tion of shapes in haptic conditions (Berthoz,1991; Millar 1981a, 1988, 1994; Paillard,1991).

Thus, the informational conditions inmemory tasks for unfamiliar tactual raised-line and raised-dot patterns are typicallycharacterised by a paucity of the referencecues that are needed for the patterns to bespatially organised as global shapes (Millar,1978 a, 1988a, 1994, 1997).

Braille patterns are an important exam-ple. The patterns lack salient features, be-cause all characters derive from a single,small (6.2 x 2 mm) matrix of six raised dots.The small size produces problems of acuity.But the principal difficulty does not lie indistinguishing patterns from each other.Discriminating characters is fairly easyeven for the inexperienced (Katz, 1925; Mi-llar, 1977a,b). The main problem lies in thepaucity of reference cues for coding the pat-terns as global shapes. The lack of redun-dancy means that there are no salient featu-res in the patterns that relate easily to eachother. That makes it difficult to code the pat-tern as a distinctive global spatial configu-ration. It is also difficult to use self-referentframes to code the pattern as a global confi-guration, because the components are alsotoo small to be related to body-centred re-ference cues to determine their position inthe pattern. In principle, the tip of the ex-ploring finger could act as a frame in rela-tion to which dot locations could be deter-mined. However, that requires some expe-rience. Inexperienced people tend to «rub»over the dots too unsystematically to bene-fit from that. External reference cues are ty-pically lacking in blind conditions, unless

MEMORY IN TOUCH


they are deliberately sought. This is in con-trast to visual conditions, which typicallyprovide concomitant reference, anchor, andupdating cues which organise visual pat-terns, including braille shapes, automati-cally as distinctive global configurations.The efficiency of visual shape recognition isreduced to the level of touch when the fieldof view is restricted so that simultaneous at-tention to the components of a pattern is nolonger possible (e.g. Loomis & Klatzky,1991). In touch, reference cues, for codingpatterns as global shapes, require explora-tion and movement. Effective exploratorystrategies are learned with experience (e.g.Berla & Butterfield, 1977).

It is thus possible to explain the evidencefor small tactual memory spans and poor re-cognition of unfamiliar tactual shapes, as adirect consequence of the paucity of re-ference information in tactual conditions,because that makes it difficult for inputs tobe spatially organised in terms of globalconfigurations (Millar, 1978 a). Global sha-pe can be regarded as an extremely econo-mical organisation of inputs to the system.Coding in terms of shape can thus performthe important function of information re-duction. The explanation harks back to Mi-ller’s (1956) contention that immediate me-mory spans are limited, and that inputs re-quire recoding in economical form, if capa-city is not to be exceeded.

Evidence that, early in learning, smallraised dot patterns tend to be coded by tex-ture or dot-density cues, rather than in termsof their global outline shape has been revie-wed extensively before (e.g. Millar, 1978a,1981a ; 1997). Briefly, the findings comefrom a number of studies that used conver-gent methods to test for shape coding intouch. Thus, generalising braille charactersto enlarged forms was found to depend onexperience with the characters. Young be-ginners needed training to recognise enlar-ged forms of braille characters that they had

learned to name. Children with more expe-rience generalised to enlarged forms wit-hout error, although it took them longer torecognise letters in enlarged form than inthe original format (Millar, 1977 a.). Errorsconsisting of missed dots are the most pre-valent (Nolan & Kederis, 1969). Outlineshape is a poor cue for braille letter recog-nition since several letters share the sameoutline. Priming patterns with connectedoutline shapes produced poorer rather thanbetter recognition, and errors in recognisingand in reproducing patterns were found todepend on failures to locate the position ofconstituent dots accurately, rather than onconfusing outline shapes (Millar, 1978a,1985a). Blindfolded sighted children foundit very easy to discriminate between braillecharacters that they had never felt before.But their drawings of four patterns that theylearned subsequently showed that they hadno idea of the shapes of the characters, sug-gesting that the global shapes of the patternshad not been the basis for their accuracy ofdiscrimination (1977 a). Matching successi-ve patterns by dot-density differences wassuperior to matching by the spatial locationof the dots, and dot-density differences pro-duced higher levels of performance thanmatching by symmetry versus asymmetry inthe shapes (Millar, 1978 a). Differences indot numerosity were also a better cue alsothan outline shape for judging the odd oneout of four braille words (Millar, 1984a).Dot-density differences were also superiorto differences in outline shape for matchingsmall dot patterns other than braille (Millar,1985 a). Recognition of braille letters wassignificantly impaired when the charactersdiffered from each other in texture (dot-gapintervals), despite the fact that they wereidentical in shape, as well as in name (Mi-llar, 1977b).

There is, of course, no question that eventhe small braille patterns that lack redun-dancy can be coded as spatial configura-

SUSANNA MILLAR


tions also by touch. That usually requiresprior experience. Direct evidence comesfrom filming the movements of the readingfinger from underneath transparent surfaces(Millar, 1988 b, 1997)). Deliberate explora-tion of characters for shape information isseen typically in very slow, letter-by-letterreaders, particularly by former print readerswith extensive visuo-spatial experience,who learned braille relatively late in life(Millar, 1997). They move the exploringfinger in up/down, zigzag and circular fas-hion over a character in attempting to cons-truct the total pattern. Such movements be-come much faster and stereotyped with ex-perience, without losing the typical charac-ter of scanning the form. For experiencedreaders, the manner of coding depends ontask demands. Fluent braillists who have le-arned braille from the start, show evidenceof coding letter shapes mainly in tasks thatdemand search for single characters. Bycontrast, fast reading for meaning is basedon cues from dynamic shear patterns acrossthe moving fingerpad (Millar, 1987). Withexperience, scanning movements by the twohands are organised, relative to each other,to provide the spatial information about thelocation of words and lines that is neededfor reading texts, as well as the tactual (she-ar pattern) information which is translatedrapidly into verbal (phonological and se-mantic) form for comprehension and recall.

Taken together, the findings support thehypothesis that the small (two to three item)tactual memory spans for unfamiliar pat-terns (see earlier) can be explained by thepaucity of reference information for codinginputs spatially in these conditions. Touchseems to be particularly good at feeling tex-ture differences. These can be used to en-hance discrimination (Millar, 1986 a ; Schiff& Isikow, 1966), and may underlie the mo-dality-specific tactual coding that was de-monstrated. Relying on texture differencesin memory when reference information is

too sparse for spatial organisation wouldcertainly be useful. But it would also ex-plain why non-verbal tactual memory spanswere confined to two to three items. Thatwas in contrast to large spans for inputs thatare quickly re-coded verbally via long-termfamiliar names that can be rehearsed in theshort term (see earlier).

The hypothesis that spatial coding de-pends on the amount and redundancy of avai-lable reference information can also accountfor better recognition of 3-D forms that arenot necessarily nameable. Such conditionsexist, for instance, for experienced subjectswho use exploratory strategies that activelyseek environmental reference cues, or use thetwo hands relative to the body midline to actas self-referent frames in relation to whichtactual «feels» can be localised.

It should be noted that the term «framesof reference» is used here as an operatio-nally defined term. The relevant informa-tion can be experimentally manipulated.Blindfolding eliminates current informationabout environmental frames in relation towhich the location of a stimulus can be de-termined. Such manipulation leaves body-centred reference frames intact. But self-re-ferent frames can also be enhanced or dis-turbed. Self-reference that is centred on thebody midline can be disrupted by changingthe posture or orientation of the body, or theorientation of displays (e.g. Millar, 1975c,1985 b). Similarly, stimuli can be aligned tofacilitate self-referent coding. We showedthat positioning the two exploring fingersdirectly above the stimuli, in alignment withthe body midaxis, produced a similar ad-vantage for symmetric over asymmetric(raised line) patterns to that which is foundin vision (Ballesteros, Millar & Reales,1998; Millar, Ballesteros & Reales, 1994).By contrast, shape symmetry is not an ef-fective cue in haptic conditions that affordfew reference cues for spatial coding (Mi-llar, 1978 a). Exploring unfamiliar shapes

MEMORY IN TOUCH


blind with one finger, without special align-ment to the body-centred frames providesfew references cues. In such conditions,symmetry has no advantage (Ballesteros,Millar & Reales, 1998, Millar, Ballesteros& Reales, 1994; Millar 1978 a).

Reference frames in relation to which thelocation of tactual features can be determi-ned are needed for spatial coding. The hy-pothesis here is that spatially organised con-figurations reduce inputs, without loss of in-formation, to a form in which the informa-tion is manageable in memory over theshort term. It predicts good memory for spa-tially organised tactual inputs in contrast torelying on relatively unorganised texture as-pects of tactual input.

Studies concerned with nonverbal tactualstimuli have mainly used single stimulusconfigurations that are larger than braillepatterns (e.g. Warm & Foulke, 1968; Locher& Simmons, 1978 ). The next section focu-ses more specifically on factors in memoryfor the somewhat larger movements thatsuch layouts require.

Short-term memory for movements

Movements have mainly been studied asoutput systems. The focus has been on thekind of control (feedback, feed-forward,open-loop or closed-loop) that accounts forcontrols in sensorimotor and motor skills.However, recent work has emphasised theimportance, as well as the diversity, of re-ference frames in relation to which move-ments are organised spatially (e.g. Berthoz,1991; Jeannerod, 1988, 1991; Paillard,1991). Most of the work has concerned mo-vements in visual environments in whichthe goal or target and other external cueswere visible, either throughout the tests, orinitially and at various points before blind-folding. A good deal of the evidence on spa-tial coding, for instance, of reaching move-ments, comes from studies with blindfolded

sighted subjects who were initially allowedsight of the target. The information subjectshave at the beginning of the task was thus ofthe target location that could, in principle,be determined in relation to surrounding ex-ternal visuo-spatial frame cues, as well asrelative to body-centred cues. When exter-nal frame cues are excluded by blindfoldingsubjects, the postural, body-centred framesthat guide reaching and aiming movementssustain memory for the target location. Intotally blind conditions, external environ-mental targets can be signalled initially byauditory or olfactory cues to which posturesare adjusted.

Short-term memory for blind movementswithin personal space (reach) has also beenstudied in the context of attentional or capa-city limitations (Laabs & Simmons, 1981).Laabs (1973) distinguished between codinglocation and movement or kinaesthetic in-formation by requiring recall of either thelocation or the extent of a positioning mo-vement, from a different position than inpresentation. In such paradigms, the endlo-cation is recalled much more accuratelythan the extent of the movement. The fin-dings were widely interpreted to suggestthat movements and spatial location are co-ded differently in short-term memory (Kel-so & Wallace, 1978; Kelso & Clarke, 1982;Laabs & Simmons, 1981; Marteniuk, 1978;Russell, 1976). Thus the endlocations ofblind movements can be coded spatially byreference to body-centred spatial frames,while recall of movement extents dependson kinaesthetic inputs when these are notorganised relative to spatial anchors or fra-mes. There has long been evidence thateven unorganised movements survive inshort-term recall. Thus, interpolating diffe-rent, irrelevant movements in delay periods,distorts recall of the target movement signi-ficantly (Adams & Djikstra, 1966).

Recall of endlocations and of movementextents of positioning movements is not

SUSANNA MILLAR


completely independent in paradigms inwhich recall starts from a different position(e.g. Walsh and Russell, 1979, 1980). Thus,subjects undershoot endlocations if the startin recall is further away, and overshoot ifthe start location is nearer than the start inpresentation. Precisely the opposite patternof over-and undershooting is found in recallof movement extent or distance from posi-tions that are further away or nearer than theoriginal starting point of the movement(Imanaka, & Abernethy, 1992 a). It has be-en suggested that subjects use a locationstrategy when location information is relia-ble, but use other means, such as trying to ti-me movements by counting, when locationis unreliable (e.g. Diewert & Roy, 1978).Explicit information about starting and end-positions of a movement enabled subjects touse a «location» strategy which abolishedthe typical over-and undershooting patterns.The typical pattern reappeared when that lo-cation was made unreliable and subjectswere asked to use counting (timing) or toenvisage the distance from a changed posi-tion about which they had no information(Imanaka & Abernethy, 1992 b). The fin-dings can be explained in terms of codingendlocation spatially, and relying on me-mory for kinaesthetic cues, depending onthe reliability of the spatial information.

External frame information can also beprovided, and its use encouraged by instruc-tions, in blind conditions. Thus deliberatelyproviding an external frame, and instructingblindfolded sighted children to use it to re-member the endlocations of positioning mo-vements from different starting points, pro-duced greater accuracy, and significantly re-duced interference from changes in thelength of movements (Millar, 1985 b). Deli-berately instructing the subjects to use theirbody midline as a spatial anchor had simi-lar, although less dramatic effects.

It has been shown that reliable spatial in-formation does not need to be visual in ori-

gin. Coding locations spatially relative toegocentric frames has been demonstratedwith young congenitally totally blind chil-dren. Indeed, young congenitally totallyblind children tend to rely more on body-centred spatial coding (Millar 1981 b, 1985b). Such spatial information is, of course,reliable also in the absence of vision. Whenself-referent cues are unreliable and thereare no salient (or well-known) external fra-me cues, they tend to use memory for mo-vements (Millar, 1975c, 1979). Using a cog-nitive interference task (counting back-wards) in conjunction with the Laabs (1973)paradigm, we found that congenitally to-tally blind children showed interference inmemory for locations that could be codedreliably by reference to body-centred fra-mes, but no interference in memory for mo-vements, when reference cues were unrelia-ble (Millar, 1994, p. 148). Memory for loca-tions that could be coded reliably by re-ference to body-centred frames showed sig-nificant interference by the difficult cogniti-ve task during delays, while such tasks didnot interfere with memory for movementsthat could not be coded reliably in terms ofspatial locations.

The findings thus demonstrate short-termmemory for movements also in the total ab-sence of visual information. But they sug-gest also that memory for movement ex-tents, coded in terms of the kinaesthetic in-puts, tends to be less accurate and more va-riable than movement information about theendlocation if these are coded spatially, forinstance, with reference to body-centredframes.

Spatial coding is not the only form of or-ganising movement information. Blind chil-dren are taught to estimate distances bycounting the number of steps, or sounds of aturning wheel. Counting during delays hasbeen found to impair kinaesthetic memory(Williams, Beaver, Spence & Rundell,1969). But the degree of practice is also im-

MEMORY IN TOUCH


portant. Constant repetition of a given mo-vement distance interfered significantlywith memory for endlocations of the move-ment when the starting position of these va-ried (Millar, 1985 b). Thus, with extensivepractice, memory for movement extents canbecome extremely robust also. The wholequestion of effects of familiarity on short-term memory spans and on working me-mory is extremely important, not least forthe recall and recognition of movement in-formation, and will be considered in moredetail later.

The point that needs to be stressed is thatshort-term memory for kinaesthetic inputscan be demonstrated even when these inputsare not spatially organised, although me-mory for unorganised kinaesthetic inputs(e.g. movement extent) is less accurate andless stable. We have shown modality-speci-fic motor memory in blind conditions fromwhich any influence from longer-term vi-sual knowledge can be excluded (Millar &Ittyerah, 1991). Recall of a criterion move-ment by congenitally totally blind childrenshowed signficant overshooting when anirrelevant larger movement was interpolatedduring the delay period, and significant un-dershooting of the criterion in recall whenthe interpolated irrelevant movement wasshorter than the criterion. Blindfolded sigh-ted children showed the same effects. But inthe case of the blind children it was alsopossible to exclude any possibility that themodality-specific effects on short-term me-mory could have been mediated by longer-term visual knowledge.

Perhaps more important still, merelyimagining the irrelevant movements duringthe delay period, without actually executingthem, had similar effects on short-term re-call of a criterion movement as actually ex-cecuting the irrelevant movement (Millar &Ittyerah, 1991). Such movement imagerywas shown also in conditions of total blind-ness. Congenitally totally blind children, as

well as blindfolded sighted children, wereinstructed to imagine («in their heads») exe-cuting an irrelevant shorter or longer move-ment during the delay before recalling thecriterion, but not to make the movementsovertly. Significant undershooting andovershooting in recall, depending on the ty-pe of irrelevant movement that had beenimagined, was shown by the congenitallytotally blind as well as by blindfolded sigh-ted children.

The finding shows movement represen-tation in short-term memory. This is an im-portant finding, because it suggests a basisfor mental rehearsal of movements even inconditions that exclude past as well as pre-sent visual information totally. Mental rehe-arsal of movements has long been shown toimprove performance by sighted adults (e.g.Johnson, 1982), and is used in sports trai-ning. The effects of imagining irrelevantmovements with blind children thus suggestthat such strategies are, in principle, alsoavailable in totally blind conditions. Ins-tructions to mentally rehearse the criterionmovement («imagine repeating the move-ment in your head») during delays signifi-cantly improved recall by the blindfoldedsighted children. The improvement was inthe same direction for the congenitally to-tally blind. But it did not reach significancelevel, despite the fact that discrepant move-ments during delays significantly interferedwith their recall (see earlier). The lack ofsignificant improvement is probably explai-ned by differences in informational condi-tions. All movements in the study had beendesigned to cross the body midline, in orderto encourage movement coding by reducingspatial coding in terms of the body midline(Millar & Ittyerah, 1991). Young blind chil-dren tend to rely on memory for movementswhen coding in terms of self-referent spatialframes is made difficult or disrupted (Mi-llar, 1979, 1981b, 1985b, 1994). Memoryfor the criterion movement would be liable

SUSANNA MILLAR


to disruption by the discrepant interpola-tions in delays if the criterion was coded interms of kinaesthetic cues. But such codingmay not be sufficiently economical to sus-tain active mental rehearsal to improve re-call. Rehearsal strategies are likely to bemore efficient with codes that organise in-puts economically (see earlier). Thus theblindfolded sighted may be able to use ad-ditional reference frames, derived from vi-suo-spatial experience, to organise kinaest-hetic inputs when self-referent spatial co-ding is made difficult. Effective mental re-hearsal of kinaesthetic cues may depend onusing more economical codes that reducethe memory load of kinaesthetic inputs.Spatial coding, either in terms of body-cen-tred or external frames, would produce bet-ter recall for that reason.

Mental rehearsal of movements can beshown to have physiological correlates.Thus physiological studies of motor ima-gery have shown that patterns of activity inbrain areas, including the motor cortex, aresimilar to patterns for actually executed mo-vements (Jeannerod & Decety, 1995). Itmay be important that subjects in most mo-vement studies are sighted people who aretested blindfolded. In some cases, testingoccurs in the dark, but an external orientingcue remains visible throughout. It seemspossible that in conditions of total blind-ness, efficient organisation of movements,which permits effective mental rehearsal,requires more practice, familiarity and con-sequently more influence of longer-termknowledge for mental rehearsal of move-ments to produce the same level of facilita-tion in short-term memory. That requiresfurther study.

The whole question of longer-term me-mory involvement in short-term and wor-king memory is extremely important, not le-ast for the recall and recognition of move-ment information. There is little doubt thatimmediate memory spans are larger for fa-

miliar than for new words, and for tactualpatterns for which names can be retrievedeasily and fast (e.g. Henry & Millar, 1991,1993; Millar, 1975a). It seems likely thatthis is true also for immediate memory formovements. Many everyday movementsare, of course, so well practiced that theyrun off automatically with great precision.A discussion of the feedback and feed-for-ward processes by which movement accu-racy is achieved is beyond the present brief.The point is rather that short-term motormemory in blind conditions can be based onkinaesthetic inputs, but that efficient mentalrehearsal of movements, which facilitatesrecall, involves longer-term memory for ef-ficient means of organising the input infor-mation so as to reduce its memory load.

The longer-term information that I haveparticularly in mind, as one basis of short-term memory for movement inputs, arepracticed exploratory movements. The im-portance of input-output links for mental re-hearsal and the size of immediate memoryspans has been mentioned before in connec-tion with memory for sounds (see, Henry &Millar, 1991, 1993). But there is also evi-dence suggesting that familiarity with well-organised exploratory movements form animportant basis of tactual recognition me-mory.

Thus, a surprising finding in getting con-genitally totally blind children to draw wasthat that even though they had never drawnbefore, the older subjects at least drew figu-res that differed little in general schemafrom those of their sighted cohorts (Millar,1975 c). More surprising still, the blind chil-dren were much better at producing recog-nisable raised line drawings of the humanfigure than at recognising such figures (Mi-llar, 1986 b, 1990, 1991 a). For the sighted,the reverse is the case. Young sighted chil-dren can recognise drawings long beforethey can attempt to reproduce them. Thefinding for the blind should not, of course,

MEMORY IN TOUCH


be taken to mean that they never recognisedrawings immediately, let alone that suchrecognition is impossible. But it does requi-re more familiarity with exploratory strate-gies than is needed in vision.

The reversal in difficulty between recog-nition and production focuses attention onthe importance of both movement output,and familiarity with exploratory strategies,as factors in short-term motor memory. Thefindings suggest strongly that familiaritywith efficiently organised exploratory out-put movements can serve as, at least one,important basis for haptic recognition andshort-term haptic memory in the total ab-sence of sight (Millar, 1991).

Intersensory coding and haptic memory

I have suggested that memory for tactualinformation has to be examined in the con-text of intersensory processing, in whichconverging inputs vary with task demands.In blind conditions, inputs for two-dimensio-nal stimulus patterns come from touch, mo-vement, and posture cues, in varying con-junctions with longer-term procedural andorganizational knowledge. The findings thathave been reviewed here suggest how infor-mation from touch and movement may becoded in short-term memory, and what infor-mational conditions produce different levelsof efficiency in recognition and recall spans.

Not surprisingly, general principles thatapply to short-term memory for inputs fromother modalities also apply to memory forhaptic inputs. One of the most important isthe principle of economy of coding (Miller,1956). Briefly, the more parsimonious theorganisation of inputs, the greater is the pro-bability of recall. The principle implies thatthere are limits to the amount of informationthat the organism can handle at any one ti-me. The fact that such «capacity limits» arenot rigid is shown by the almost universaleffect of familiarity and repetition. Familia-

rity effects are also (though not only) im-portant in the modes of recoding, or reorga-nizing inputs more parsimoniously, that areavailable to a subject. The second principleis, therefore, that tasks, which require tem-porary memory, also involve longer-termmemory to greater or lesser extents.

The third principle, advocated here, isthat limits on short-term functioning dependon the amount of overlap and redundancy ofconverging information from different sour-ces. Neither the metaphor of a single limi-ted-capacity channel, nor the notion of num-bers of quite separate (floppy-disk-like) mo-dules, are adequate. The alternative metap-hor is needed particularly for increasing evi-dence on the relation between the modali-ties. Findings suggest that high degrees ofspecialization, on the one hand, but alsohigh degrees of informational overlap, onthe other hand, are needed to organise in-puts spatially in terms of reference frames.

Multisensory information is indeed thenorm for humans in most conditions. It isinconceivable, on general grounds, thatmultiple specialized mechanisms would ha-ve evolved if each provided precisely thesame higher order abstract information.Specialized series of analyses of inputsfrom different external and internal sourcesconverge in varying combinations to provi-de multisensory information. Thus, the phy-siological evidence shows that a number ofareas of the brain that subserve spatial tasksreceive multisensory information (e.g.Stein, 1991). Single unit recording hasfound bimodal neurons, which are activatedboth by visual and tactual stimuli, in a num-ber of brain areas which are involved in re-aching movements, suggesting how spatialorganisation of reaching movements may beencoded relative to extrapersonal space(Graziano & Gross, 1994, 1995). Multisen-sory information and overlap seems to beparticularly important for the spatial organi-sation of inputs.

SUSANNA MILLAR


Experimental studies of crossmodal, vi-sual/tactual performance have mainly usedparadigms which dissociate the contributingmodalities in spatial tasks. There is ampleevidence from these that viewing the handthrough distorting lenses alters subjects’haptic perception of where their hands are(e.g. Howard & Templeton, 1966). Cross-modal coding of 3-D shapes has been de-monstrated with young children (e.g. Rudel& Teuber, 1964). But it involves more thanone factor. Task conditions, even the orderin which intramodal and crossmodal condi-tions are presented, have signficant effects(e.g. Millar, 1975 c). Fewer studies actuallyuse simultaneous bimodal inputs. But it isfairly clear from the findings that concomi-tant inputs from different sources facilitateperformance particularly in conditions inwhich the available task information is de-graded or insufficient. Thus adding infor-mation from touch to vision contributes lit-tle to bettering visual shape recognition, butadding visual information aids recognitionof unfamiliar tactual shapes (Millar, 1971;Heller, 1982). Similarly, concordant inputsfrom texture and shape cues improve tactualperformance, while discordant inputs dis-rupt recognition (Millar, 1986 c). The tasksthat are of special interest here concern re-cognition and memory of relatively unfami-liar 2-D shapes and displays. Such tasks arerarely a problem in normal visual conditionseven for very young children (e.g. Balleste-ros et al, 1998; Millar, 1971, 1972a; Millaret al, 1994). Visual conditions typically pro-vide the concomitant cues from differentsurfaces and features that act as referenceframes for spatial coding from the start. Mo-reover, they usually overlap with conver-gent proprioceptive reference cues.

The same tasks for which vision providesthe most salient concomitant reference cuescan also be performed with purely haptic in-formation. But the automatic overlap andredundancy of concomitant current referen-

ce cues from different sources is severelyreduced the absence of sight. It is possible,in principle, to achieve the same levels ofefficiency in haptic as in visual tasks that re-quire memory for 2-D shapes, distances, di-rections, or locations, as in vision, providedthat alternative means of reference are avai-lable for spatial coding. At the same time,information conditions are not precisely thesame for blindfolded sighted subjects, as forthe congenitally totally blind who have nolong-term visual experience (Millar, 1979,1988 a, 1994). Body-centred reference fra-mes, prior procedural experience in sear-ching for external frame cues, and cue re-dundancy from alternative (e.g. hearing)sources become more important for spatialcoding in these conditions.

Memory for information from touch andmovement, both in short-term blindfolding,and in the total absence of prior visual ex-perience, has been demonstrated. The prin-ciple of parsimony applies. Short-termspans for unfamiliar inputs are small. Butthey show effects of coding texture (shearpattern) in the case of shape tasks, and kina-esthetic coding in the case of unfamiliarmovements, or both. Such memory codingis «modality-specific» in the sense that co-ding is derived from relatively specific as-pects of the input.

More robust short-term haptic memory isfound when the same haptic inputs are spa-tially organised. Such coding depends onaccessible reference frame cues, as doesspatial coding in visual conditions. The sa-me general principle thus applies. Neverthe-less, the informational conditions on whichsuch coding depends differ from those inwhich visual cues are also present. Thusthere is greater need to rely on body-centredreference information, or prior proceduralknowledge, or to use alternative external(e.g. sound) location cues when that is pos-sible. Alternative means of more parsimo-nious coding can involve naming (see ear-

MEMORY IN TOUCH


lier), or counting. Moreover, the findingsshow that short-term haptic memory can in-volve mental rehearsal of movements. Thisdiffers from the articulatory coding that in-creases verbal memory spans. Memoryeven for the more organised haptic inputsthus includes information that derives fromthe haptic input conditions. Memory intouch thus includes «modality-specific» as-pects of the input information.

To fit these findings into some form of«working memory» model, one might adda haptic - movement loop system, in ana-logy with articulatory coding. But to workeffectively, the system would also need ac-

cess to longer-term visuo-spatial and/oregocentric reference information, as wellas to longer-term procedural knowledge.Such access would need to be quite flexi-ble, especially in response to differences intask demands. It is possible that the as-sumption of a «central executive» in thesystem (Baddeley, 1990) might be suffi-cient to fulfill that role. But it is not quiteclear how the model incorporates the con-tinual changes in longer-term memory thatmust be supposed with development andfurther experience, and which clearly playan important role, particularly in hapticshort-term memory.

SUSANNA MILLAR


References

Adams, J.A. & Dijkstra, S. (1966). Short-term memory for motor responses. Journal ofExperimental Psychology, 71, 314-318.

Appelle, S. (1991). Haptic perception ofform: Activity and stimulus attributes. In Heller& W. Schiff (Eds) The Psychology of Touch (pp169-188). Hillsdale, N.J.: Erlbaum.

Anderson, R.A. (1987). Inferior parietal lobu-le function in spatial perception and visuomotorintegration. In Plum & V.B. Mountcastle (Eds)Handbook of Physiology. Rockville, Maryland:American Physiological Society.

Atkinson, R.C. & Shiffrin, R.M. (1968). Hu-man Memory: A Proposed System and its Con-trol Processes. In K. Spence & J.T. Spence (Eds),The Psychology of Learning and Motivation,Vol.2. London: Academic Press.

Atwood, G. (1971). An experimental study ofof visual imagination and memory. CognitivePsychology, 2, 290-299.

Baddeley, A.D. (1986). Working Memory.Oxford: The Clarendon press.

Baddeley, A.D. (1990). Human Memory:Theory and Practice. Hove: Lawrence ErlbaumAssociates.

Baddeley, A.D. & Hitch, G. (1974). Workingmemory. In G. Bower (Ed). The Psychology ofLearning and Motivation. Vol. VIII, pp. 47-89.New York: Academic Press.

Ballesteros, S. Millar, S. & Reales, J.M.(1998). Symmetry in haptic and in visual percep-tion. Perception & Psychophysics, 60, 389-404.

Berla, E.P. & Butterfield, L.H.Jr. (1977). Tac-tual distinctive feature analysis: Training blindstudents in shape recognition and in locating sha-pes on a map. The Journal of Special Education,11, 336-346.

Berthoz, A. (1991). Reference frames for theperception and control of movement. In J. Pai-llard (Ed). Brain and Space. Oxford: OxfordUniversity Press.

Besner, D. (1987). Phonology, lexical accessin reading, and articulatory suppression: A criti-cal review. Quarterly Journal of ExperimentalPsychology, 41A, 91-105.

Broadbent, D. (1984). The maltese cross: anew simplistic model for memory. Behaviouraland Brain Sciences, 7, 55-94.

Brooks, L.R. (1968). Spatial and verbal com-ponents of the act of recall. Canadian Journal ofPsychology, 22, 349-368.

Colley, A. & Colley, M. (1981). Reproductionof endlocation and distance of movement inearly and later blind subjects. Journal of MotorBehavior, 13, 102-109.

Conrad, R. (1964). Acoustic confusions inimmediate memory. British Journal of Psycho-logy, 55, 75-84.

Conrad, R. (1971). The chronology of the de-velopment of covert speech in children. Deve-lopmental Psychology, 5, 398-405.

Davidson, P.W. (1972). The role of explora-tory activity in haptic perception: Some issues,data and hypotheses. Research Bulletin, Ameri-can Foundation for the Blind, 24, 21-28.

Davidson, P.W. (1974). Some functions of ac-tive handling: Studies with blinded humans. NewOutlook for the Blind, 70 (5), 198-202.

Davidson, P.W., Barnes, J.K. & Mullen, G.(1974). Differential effects of task memory de-mands on haptic matching of shape by blind andsighted humans. Neuropsychologia, 12, 395-397.

Diewert, G.L. & Roy , E.A. (1978). Codingstrategy for movement and extent information.Journal of Experimental psychology: Human:Learning & Memory, 4, 666-675.

Dobelle, W.H., Mladejovsky, J.P. & Girvin,J.P. (1974). Artificial vision for the blind: Elec-trical stimulation of the visual cortex offers hopefor a functional prothesis. Science, 183, 440-444.

Fodor, J. (1983). Modularity of Mind. Cam-bridge, Mass.: MIT Press.

Gibson, J.J. (1962). Observations on activetouch. Psychological Review, 69, 477-491.

Gibson, J.J. (1966). The Senses Consideredas Perceptual Systems. Boston: HoughtonMiflin.

Gibson, J.J. (1979). The Ecological Approachto Visual Perception. Boston: Houghton Mifflin.

Gilson, E.Q. & Baddeley, A.D. (1969). Tacti-le short-term memory. Quarterly Journal of Ex-perimental Psychology, 21, 180-189.

Goldenberg, G., Podreka, I., & Steiner, M.(1990). The cerebral localization of visual ima-gery: evidence from emission computerized to-mography of cerebral blood flow. In P.J.Hamp-son, D. Marks, & J.T.E. Richardson (Eds) Ima-gery: Current Developments. London: Routledge.

Goodnow, J.J. (1971). Eye and hand: diffe-rential memory and its effect on matching.Neu-ropsychologia, 9, 89-95.

Graziano, M.S.A. & Gross, C.G. (1994). Therepresentation of extrapwersonal space: A possi-ble role for bimodal visual-tactile neurons. In Gaz-zaniga, M.S. (Ed) The Cognitive Neurosciences(pp. 1.021-1.034.). Cambridge, Mass: MIT Press.

Graziano, M.S.A. & Gross, C.G. (1995).Mapping space with neurons. Current Directionsin Psychological Science, 3 (5), 164-167.

Haber , R.N. & Haber, R.B. (1964). Eideticimagery: I Frequency. Perceptual & MotorSkills, 19, 131-138.

Hatwell, Y. (1978). Form perception and rela-ted issues in blind humans. In R. Held,

H.W. Leibowitz & H.L.Teuber (Eds) Hand-book of Sensory Physiology. Berlin: SpringerVerlag.

Henry, L. & Millar, S. (1991): Memory spanincrease with age: A test of two hypotheses.Journal of Experimental Child Psychology. 51,458-484.

Henry, L.A. & Millar, S. (1993). Why doesmemory span improve with age? A review of theevidence for two current hypotheses. EuropeanJournal of Cognitive Psychology, 5, 241-287.

Heller, M. A. (1982). Visual and tactual tex-ture perception: Intersensory cooperation. Per-ception & Psychophysics, 31, 339-344.

Hitch, G. J. & Halliday, M.S. (1983). Workingmemory in children. Philosophical Transactionsof the Royal Society of London, B302, 325-340.

Hitch, G.J., Halliday, M.S. Schaafstal, A.M.& Schraagen, J.M.L. (1988). Visual working me-mory in young children. Memory & Cognition,16, 120-132.

Howard, I . P. & Templeton, W.B. (1966). Hu-man Spatial Orientation. New York: John Wiley.

Hulme, C. & Tordoff, V. (1989). Working me-mory development: The effects of speech rate,word length , and acoustic similarity on serial re-call. Journal of Experimental Child Psychology,47, 72-87.

Imanaka, K. & Abernethy, B. (1992 a). Inter-ference between location and distance informa-tion in short-term motor memory: The respectiveroles of kinaesthetic signals and abstract codes.Journal of Motor Behavior, 24, 274-280.

Imanaka, K. & Abernethy, B. (1992 b). Cog-nitive strategies and short-term memory for mo-vement distance and location. Quarterly Journalof Experimental Psychology, 45A (4) 669-700.

Jeannerod, M. (1988). The Neural and Beha-vioural Organization of Goal-directed Move-ments. Oxford: Clarendon Press.

Jeannerod, M. (1991). A neurophysiologicalmodel for the directional coding of reaching mo-vements. In J. Paillard (ed.) Brain and Space.Oxford: Oxford University Press.

Jeannerod, M. & Decety, J. (1995). Mentalmotor imagery: a window into the representatio-nal stages of action. Current Opinion in Neuro-biology, 5, 727-732.

Johnson, P. (1982). Functional equivalence ofimagery and movement. Quarterly Journal ofExperimental Psychology, 34A, 349-365.

MEMORY IN TOUCH


Katz, D. (1925). Der Aufbau der Tastwelt.Leipzig: Barth.

Klatzky, R.L., Lederman, S.J. & Metzger,V.A. (1985). Identifying objects by touch. Per-ception & Psychophysics, 37, 299-302.

Kelso, J.A.S. & Wallace, S.A. (1978). Cons-cious mechanisms in movement. In. G.E. Stel-mach (Ed) Information Processing in MotorControl and Learning. New York: AcademicPress.

Kelso, J.A.S. & Clarke, J.E. (1982). The De-velopment of Movement Control. New York:John Wiley.

Kosslyn, S.M. (1980). Image and Mind. Har-vard: Harvard University Press.

Kosslyn, S.M. (1981). The medium and themessage in mental imagery: A theory. Psycholo-gical Review, 88, 46-66.

Kosslyn, S.M., Ball, T.M., & Reiser, B.J.(1978). Visual images preserve metric spatial in-formation: Evidence from studies of image scan-ning. Journal of Experimental Psychology: Hu-man Perception & Performance, 4, 47-60.

Laabs, G.L. (1973). Retention characteristicsof different reproduction cues in short-term me-mory. Journal of Experimental Psychology, 100,168-177.

Laabs, G.L. & Simmons, R.W. (1981). Motormemory. In D. Holding (Ed) Human Skills. NewYork: Wiley.

Locher, P. J. & Simmons, R.W. (1978). In-fluence of stimulus symmetry and complexityupon haptic scanning strategies during detection,learning and recognition tasks. Perception &Psychophysics, 23, 110-116

Logie, R.H. (1986). Visuo-spatial processingin working memory. Quarterly Journal of Expe-rimental Psychology, 38A, 229-247.

Logie & Baddeley, A.D. (1990). Imagery andworking memory. In P.J.Hampson, D.F. Marks,& J.T.E. Richardson (Eds) Imagery: Current De-velopments. London: Routledge.

Logie, R.H. & Pearson, D.G. (1997). The in-ner eye and the inner scribe of visuo-spatial wor-king memory: Evidence from developmentalfractionation. European Journal of CognitivePsychology, 9 (3), 241-257.

Logie , R.H., Zucco, G.M. & Baddeley, A.D.(1990). Interference with visual short-term me-mory. Acta Psychologica, 75, 55-74.

Loomis, J.M. & Klatzky, R.L. (1991). Simila-rity of tactual and visual picture recognition withlimited field of view. Perception, 20, 167-177.

Marteniuk, R.G. (1978). The role of eye andhead position in slow movement execution.InG.E. Stelmach (Ed). Information processing inMotor Control and Learning. New York: Acade-mic Press.

Millar, S. (1971). Visual and haptic cue utili-zation by preschool children: The recognition ofvisual and haptic stimuli presented separatelyand together. Journal of Experimental Child Psy-chology, 12, 88-94.

(1972a). Effects of instructions to visualisestimuli during delay on visual recognition bypreschool children. Child Development, 43,1.073-75.

(1974). Tactile short-term memory by blindand sighted children. British Journal of Psycho-logy, 65, 253-263.

(1975a). Effects of tactual and phonologicalsimilarity on the recall of Braille letters by blindchildren. British Journal of Psychology, 66, 193-201.

(1975b). Effects of phonological and tactualsimilarity on serial object recall by blind andsighted children. Cortex , 11, 170-180.

(1975 c). Spatial memory by blind and sigh-ted children. British Journal of Psychology, 66,449-459.

(1975 d). Translation rules or visual experien-ce? Drawing the human figure by blind and sigh-ted children. Perception, 4, 363-371.

(1975 e). Effects of input variables on visualand kinaesthetic matching by children within andacross modalities. Journal of Experimental ChildPsychology, 1975, 19, 63-78.

(1977 a). Early stages of tactual matching.Perception, 6, 333-343.

(1977 b). Tactual and name matching by blindchildren. British Journal of Psychology, 68, 377-387.

(1978 a). Aspects of information from touchand movement. In G. Gordon (ed.) Active Touch.London: Pergamon Press.

(1978 b). Short-term serial tactual recall: Ef-fects of grouping tactually probed recall of Brai-lle letters and nonsense shapes by blind children.British Journal of Psychology, 69, 17-24.

(1979). Utilization of shape and movementcues in simple spatial tasks by blind and sightedchildren. Perception, 1979, 8, 11-20.

(1981 a). Crossmodal and intersensory per-ception and the blind. In Walk & Picks (eds.) In-tersensory Perception and Sensory Integration.Oxford:, Plenum Press. 34.

SUSANNA MILLAR


(1981 b). Self-referent and movement cues incoding spatial location by blind and sighted chil-dren. Perception, 10, 255-264.

(1984). Strategy choices by young Braille re-aders. Perception, 13, 567-579.

(1985 a). The perception of complex patternsby touch. Perception, 14, 293-303.

(1985 b). Movement cues and body orienta-tion in recall of location by blind and sightedchildren. Quarterly Journal of ExperimentalPsychology, 37, 257-279.

(1986 a). Aspects of size, shape and textu-re in touch: Redundancy and interference in chil-dren’s discrimination of raised dot patterns.Journal of Child Psychology and Psychiatry, 27,367-381.

(1986 b). Drawing as Image and Representa-tion in Blind Children. In D.G. Russell, Marks &T.E. Richardson (Eds) Image 2. Dunedin, NewZealand:

(1987 a). Perceptual and task factors in fluentbraille. Perception, 1987, 16, 521-536.

(1988 a). Models of Sensory Deprivation:The nature/nurture dichotomy and spatial repre-sentation in the blind. International Journal ofBehavioural Development, 11, 69-87.

(1988 b). An apparatus for recording handmovements. British Journal of Visual Impair-ment and Blindness, VI, 87-90.

(1990) Imagery and Blindness. In Hampson,D.F. Marks & J.T.E. Richardson (Eds) Imagery:Current Developments. London: Routledge &Kegan Paul.

(1991). A Reversed Lag in the Recognitionand Production of Tactual Drawings: TheoreticalImplications for Haptic Coding. M.A. Heller andW. Schiff (Eds) The Psychology of Touch. N.J.:Lawrence Erlbaum Associates.

(1994). Understanding and Representing Spa-ce: Theory and Evidence from Studies withBlind and Sighted Children. Oxford: ClarendonPress.

(1997). Reading by Touch. London & N.Y.:Routledge.

Millar, S. Ballesteros & J. M. Reales (1994).Influence of symmetry in haptic and visual per-ception, Paper presented at the 35th Annual Me-eting of the Psychonomics Society, St. Louis,Mo. November 11-13.

Millar, S. & Ittyerah, M. (1991). Mental prac-tice without visuo-spatial information. Interna-tional Journal of Behavioral Development, 15,125-146.

Miller, G. (A. 1956). The magical number se-ven plus or minus two: Some limits on capacityfor processing information. Psychological Re-view, 63, 81-87.

Monsell, S. (1984). Components of workingmemory underlying verbal skills: A «distributed ca-pacities» view. In H. Bouma & D. Bouwhuis (Eds)International Symposium on Attention and Perfor-mance, X (327-350). Hillsdale, N.J.: Erlbaum.

Monsell, S. (1987). On the relation betweenlexical input and output pathways for speech. InA. Allport, D.G.Mackay, W.Prinz, & E. Scheerer(Eds) Language Perception and Production: Re-lations between Listening, Speaking, Readingand Writing. London: Academic Press.

Nolan, C., Y., & Kedeeris, C.J. (1969). Per-ceptual Factors in Braille Recognition. AmericanFoundation for the Blind Research Series, No.23.New York: American Foundation for the Blind.

Paillard, J. (1991). Motor and representatio-nal framing of space. In J. Paillard (Ed) Brainand Space. Oxford: Oxford University Press.

Revesz, G. (1950). Psychology and Art of theBlind. London: Longmans.

Rudel, R.G. & Teuber, H.L. (1964). Crossmo-dal transfer of shape information by children.Neuropsychologia, 2, 1-8.

Russell, D.G. (1976). Spatial location cuesand movement production. In G. Stelmach (Ed)Motor Control: Issues and Trends. New York:Academic Press.

Sakata, H. & Iwamura, Y. (1978).Corticalprocessing of tactile information in the first so-mato-sensory and parietal association areas inthe monkey. In G. Gordon (Ed) Active Touch.New York: Pergamon Press.

Schiff, W. & Isikow, H. (1966). Stimulus re-dundancy in the tactile perception of histograms.International Journal of Education for the Blind,16, 1-10.

Schneider, G.E. (1967). Contrasting visuo-mo-tor functions of tectum and cortex in the goldenhamster. Psychologische Forschungen, 31, 52-62.

Segal, S.J. & Fusella, V. (1970). Influence ofimagined pictures and sounds on detection of vi-sual an auditory signals. Journal of ExperimentalPsycholog, 83, 458-464.

Shephard, R.N. & Cooper, L.A. (1982). Men-tal images and their Transformations. Cambrid-ge, Mass: MIT Press.

Shephard, R.N & Feng, C. (1972). A chrono-metric study of mental paperfolding. CognitivePsychology, 3, 228-243.

MEMORY IN TOUCH


Shimizu, Y, Saida, Sh. & Shimura, H. (1993).Tactile pattern recognition by graphic display:Importance of 3-D information for haptic per-ception of familiar objects. Perception & Psy-chophysics, 53, (1), 43-48.

Simmons, R. & Locher, P. (1979). Role of ex-tended haptic experience upon perception ofnonrepresentational shapes. Perceptual & MotorSkills, 48, 987-991.

Squire, L.R. (1987). Memory and Brain. Ox-ford: Oxford University Press.

Stein, J. F. (1991). Space and the parietal as-sociation areas. In J. Paillard (Ed) Brain andSpace. Oxford: Oxford University Press.

Vallar, G & Baddeley, A.D. (1982). Short-term forgetting and the articulatory loop. Quar-terly Journal of Experimental Psychology, 34A,530- 560.

Walsh, W.D. & Russell, D.G. (1979). Me-mory for movement location and distance:Star-ting position and retention interval effects. Jour-nal of Human Movement Studies, 5, 68-76.

Walsh, W.D. & Russell, D.G. (1980). Me-mory for preselected slow movements: Evidencefor integration of location and distance. Journalof Human Movement Studies, 6, 95-105.

Warm, J. & Foulke, E. (1968). Effects oforientation and redundancy on the tactual per-ception of forms. Perceptual & Motor Skills, 27,83-89.

Watkins, M.J. & Watkins, O.C. (1974). A tac-tile suffix effect. Memory & Cognition, 5, 529-534

Waugh, N. & Norman, D.A. (1965). Primarymemory. Psychological Review, 72, 89-104.

Weber, E.H. (1834). De Tactu (translated1978, by H.E. Ross). London AcademicPress.

Weiskrantz, L. (1986). Blindsight: A CaseStudy and its Implications. Oxford: Oxford Uni-versity Press.

Williams, H.L. Beaver, W.S., Spence, M.T. &Rundell, Q. (1969). Digital and kinaesthetic me-mory with interpolated information processing.Journal of Experimental Psychology, 80, 530-536.

Wilkinson, J.M. & Carr, T.H. (1987). Strate-gic hand use preferences and hemispheric spe-cialisation in tactual reading. Brain & Language,32(1) 97-123.

Aceptado el 11 de marzo de 1999

SUSANNA MILLAR


MEMORY IN TOUCH - Psicothema

Documents

Transcript of MEMORY IN TOUCH - Psicothema