Handedness dependency in recall from everyday memory

zyxwvutsr zyxwvutsrq zyxwvut z British ]wrna/ of Psychology (1997), 88, 609-619 609 Printed in Great Britain 0 1997 The British Psychological Society Handedness dependency in recall from everyday memory zyxwvutsrq Gregory V. Jones* Department of Psychology, University of Wanvick, Coventy CV4 7AL, UK Maryanne Martin Department of Experimental Psychology, University of Oxford, UK A number of previous studies have demonstrated systematic misremembering of the direction in which the Queen’s head faces on British coins. Two experiments were carried out to investigate whether this phenomenon is affected by a person’s handedness. In both experiments, right-handed and left-handed participants were found to differ significantly in their verbal responses, with recall performance significantly worse than chance for right-handed but not for left-handed participants. Experiment 2 also examined degrees of handedness, and found significant variation in recall across the handedness range. Performance in this everyday-memory paradigm appears to be determined by both handedness and schema factors. It is proposed that although in this task the response was a verbal one, relevant motor imagery may nevertheless have been activated and led to the highly unusual observation of an effect of handedness upon cognitive performance. zy To what extent does a person’s handedness influence cognitive processing? The general answer provided by experimental studies seems to be either not at all or, at most, very little. In a review of possible cognitive correlates of handedness, Bishop (1990, p. 87) concluded by quoting an earlier review by Marshall (1981) : “‘All in all, there seems to be little cause at the moment to give up the hypothesis that lefthanders differ from right-handers solely by virtue of preferring their left hands” (p. 78). There is little in the more recent research literature to alter that conclusion’. The area of cognitive performance which has been most studied in the context of handedness is that of reading. In particular, commencing with Orton (1925), considerable research has investigated the hypothesis of a relation between handedness and developmental disorders of reading. Eglinton & Annett (1994) reanalysed the evidence in this area which was reviewed by Bishop (1990). Using the meta-analytic procedures described by Rosenthal (1991), they concluded that this evidence in fact displays a small but significant association between left-handedness and reading disorder. Are there other associations with handedness? In an influential proposal, Geschwind and his colleagues linked handedness and reading disorder to a wide range of other variables (e.g. Geschwind & Behan, 1982; Geschwind & * Requests for reprints. 610 zyxwvutsr zyxw zyx G. V.Jones and Matyanne Martin Galaburda, 1987). A comprehensive review (Bryden, McManus & Bulman-Fleming, 1994a) has concluded, however, that this wider model is not well supported by empirical evidence. In the case of memory, handedness has generally been assumed to be unimportant. A recent report by McKelvie & Aikins (1993) that this is not the case is therefore of considerable interest. They employed an everyday memory paradigm that had been used by a number of other investigators without any evidence on the issue of handedness being reported (e.g. Jones, 1990; Jones & Martin, 1992; Martin & Jones, 1992; Nickerson & Adams, 1979; Richardson, 1992, 1993; Rubin & Kontis, 1983). The paradigm originated with Nickerson & Adams (1979), who demonstrated that the recall of information which their population had encountered on innumerable occasions in their everyday life-the design of the US cent-was surprisingly poor. In similar studies of memory for British coins, it was not only confirmed that the general level of recall was poor but also systematic misremembering of coin head orientation was observed (Jones, 1990; Jones & Martin, 1992; Richardson, 1992). The Queen’s head on all British coins faces to the right, but the majority of British people tested are prone to the mnemonic illusion that it faces to the left-the proportion of the population who recall the direction correctly is reliably less than even the 50 per cent level to be expected by chance. Following Rubin & Kontis (1983), we hypothesized that people recall coin designs on the basis of a generic schema for coins. For British people, it was further proposed that the part of the schema specifying head direction is distorted by input from postage stamps (and possibly banknotes), which depict the Queen’s head facing to the left. In Canada, the Queen’s head appears on coins in the same way as in Britain, but is absent from postage stamps. Thus the coin-stamp schema explanation predicts that in Canada interference from postage stamps should be absent and the level of recall of coin head orientation should always be 50 per cent or more. This prediction was examined by Jones & Martin (1992), who found that the proportion of correct recall of coin head orientation in Canada was indeed numerically above 50 per cent, though not significantly so. Further results from Canada were reported by McKelvie & Aikins (1993). What is of particular interest for the present purposes is that in two experiments they tested equal-sized groups of left-handed and right-handed participants (Jones & Martin did not select on the basis of handedness). In their Expt 2 (sample 1) they tested two groups of 25 each and found that for left-handers and right-handers the percentages of correct recall were 72 per cent (18/25) and 52 per cent (13/25), respectively. In their Expt 3 they tested two groups of 38 each and the corresponding percentages were 50 (19/38) and 28.9 per cent (11/38), respectively. McKelvie & Aikins reported that in neither case alone did the difference between left-handers and right-handers reach significance, but that if the results of the two different experiments were combined then significance was reached: 58.7 per cent of left-handers (37/63) and 38.1 per cent of right-handers (24l/63) were correct. It should be stressed that all participants gave a verbal response as to whether the coin head pointed to the left or to the right, rather than attempting to draw the head. Thus the locus of the effect zyxw This frequency is incorrectly given as 26 on p. 361 of McKelvie & Aikins (1993). Handedness dependency in recall zyxw 611 appears to be central and cognitive rather than the consequence of peripheral graphical factors (cf. Shanon, 1979). Is the apparent effect of handedness a reliable one? It was obtained only by combining data from two experiments with substantially differing overall levels of correct performance and different procedures. Thus the overall level of correct performance was 62 per cent (31/50) in Expt 2 (sample 1) and 39.5 per cent (30/76) in Expt 3; and participants in Expt 3, unlike in Expt 2 (sample l), were asked to imagine holding a coin in the hand before they made a response. If it were a reliable finding, it would possess considerable general interest as demonstrating a hitherto unsuspected cognitive effect of handedness. It would also call into question the adequacy of the schema explanation of the mnemonic illusion observed with British participants, since there is no inherent reason to expect that the coin schemas of righthanders should be more susceptible to distorting input from stamps than those of left-handers. zyxwvut zy EXPERIMENT 1 In the light of the preceding considerations, it was decided to study recall concerning British coins by separate groups of left-handed and right-handed people. Recall concerning British postage stamps was included for comparison. Group size was set at 80, as compared with the combined size of 63 employed by McKelvie & Aikins (1993); 80 was also the group size used by Shanon (1979) and by Bishop (1984) in the analysis of cognitive test data from the National Child Development Study. Assuming that left-handed people constitute of the order of 10 per cent of the population, this implied that a population reservoir of the order of 800 or more people would need to be tapped to obtain the desired number of left-handed volunteers. For his study, Shanon (1979, p. 458) noted that ‘several thousand subjects had to be interviewed over a span of 4 years’. Method Participants zyxw zyxwv There were 160 participants, 80 left-handed and 80 right-handed (the criterion for handedness is described in the Procedure). All participants were undergraduates of the University of Oxford, with an average age of 20.0 years (SD 1.2). The proportion of females was 42 per cent overall and did not differ significantly between left-handers and right-handers (~‘(1,N = 160) = 1.49) (the unmodified Pearson statistic was employed throughout). All participants were volunteers. The participants were recruited primarily by approaching students observed to be writing with the left hand or with the right hand in university libraries. Materials All British coins in circulation at the time of testing have been issued since the start of the current reign in 1953 and bear the profile head of Queen Elizabeth I1 facing to the viewer’s right (for a general review of factors influencing coin design, see Bruce, 1989). In contrast, all standard issues of British postage stamps during this period have borne the head of the Queen facing to the left. 612 zyxwvuts zyxw zyxwvuts G . V .Jones and Maryanne Martin Procedure Participants were tested individually. Memory was first probed with the following question. ‘British coins have the head of the Queen on them. Does the Queen’s head face to your left or to your right?’ This was then followed by another question. ‘British stamps also have the head of the Queen on them. Does the Queen’s head face to your left or to your right?’ After the memory probes, handedness was elicited as follows. ‘Would you describe yourself as lefthanded or right-handed? In particular, which hand do you use for drawing?’ zyxw zyxwv z Results The proportion of left-handers who correctly recalled coin head orientation as facing to the right was 46.3 per cent (37/80), which did not differ significantly from the chance level of 50 per cent (~‘(1,N = 80) = 0.45). The same proportion for right-handers was 26.3 per cent (21/80), which was significantly below the 50 per cent level (~‘(1,N = 80) = 18.05, p < .OOl). Most importantly, the proportion correct for left-handers was significantly greater than that for right-handers (1,N = 160) = 6.92, p < .Ol). Turning to stamps, the proportions of left-handers and of right-handers ‘who correctly recalled head orientation as facing to the left were 65.0 per cent (52/80) and 73.8 per cent (59/80), respectively. Both proportions were significantly above the 50 , 80) = 7.20, p < .01 and x’(1,N = 80) = 18.05, p < .001, per cent level ( ~ ~ (N1 = respectively). There was no significant difference between the two proportions (1,N = 160) = 1.44). (x’ (x’ Discussion The first major conclusion to be drawn from this experiment is that it is possible to detect a reliable, and indeed substantial, effect of handedness upon memory performance. In recalling coin head orientation, accuracy was significantly higher for left-handed than for right-handed participants. The magnitude of the left-handed advantage observed here was, at 20.0 per cent (16/80), similar to the overall advantage of 20.6 per cent (13/63) reported previously-by McKelvie & Aikins (1993). The second major conclusion to be drawn concerns the coin-head mnemonic illusion to which it had previously been thought British populations in general were subject (e.g. Jones, 1990; Richardson, 1992, 1993). It now appears that it is only the right-handed population for which a significant majority misremember the direction in which the head on all British coins faces. What are the implications for our understanding of this mnemonic illusion? They appear to demonstrate that there are at least two factors which are operative. The first concerns handedness, as already noted. The second concerns the schema. The overall level of correct recall observed in the study of McKelvie & Aikins (1993) was 48.4 per cent (61/126) whereas the corresponding level in the present experiment was 36.3 per cent (58/160). The 12.1 per cent deterioration in the British level of performance can be attributed to the effects of schema distortion as a consequence of exposure to stamp heads facing in the opposite direction to coin heads. The most extreme levels of misremembering are thus observed with a right- zyxw zyxw Handedness dependency in recall 613 handed British population, with 73.8 per cent (59/80) incorrectly recalling that the head on a coin faces left in the present study. Are the handedness and schema factors independent in their effects? If they are, then an effect of handedness should be obtained for recall of stamp head orientation which is similar to that for recall of coin head orientation in spite of the overall higher level of accuracy, 69.4 per cent (111/160), for stamps. In the present study, the degree to which the proportion of left-handed participants exceeded right-handed ones in responding that the head faces to the right was 20.0 per cent (16/80) for coins and 8.8 per cent (7/80) for stamps. It would probably be premature to conclude that the handedness and schema factors are not independent on the basis of the numerical difference between these two proportions, given the relatively small numbers involved in this comparison. The reliability of the handedness effect in coin head recall suggests that the effect may be used to explore further the dependency of cognitive processing upon handedness. In Expt 1, participants were classified as left-handed or right-handed solely on the basis of the hand which they reported preferring for drawing. There may be, however, a spectrum of handedness between that of the pure left-hander and that of the pure right-hander. It is possible that coin head recall is sufficiently sensitive to handedness that systematic variation in recall can be detected across a range of handedness more differentiated than the simple dichotomy between lefthanded and right-handed. In order to investigate this possibility, a further experiment was carried out in which participants completed the handedness questionnaire of Annett (1970, 1985). In this questionnaire, the hand preference for each of 12 activities is recorded as ‘left’, ‘either’ or ‘right’. Following Annett, two procedures can then be used to determine degree of handedness. In the first procedure, a person is first categorized as overall left-handed or right-handed on the basis of the response concerning writing, and the other 11 responses are then used to determine whether the person’s handedness is ‘pure’ or ‘mixed’. A person’s handedness is pure if each of the other 11 responses is the same as the writing response or else is the ‘either’ response ; otherwise the person’s handedness is mixed. The second procedure obtains a more continuous measure of handedness by adding all 12 responses together. A score of 1, 2 or 3, respectively, is allocated to a right, equal or left response, yielding a handedness index with range 12 (complete right) to 36 (complete left). The additional experiment also provides an opportunity to examine further the recall of stamp head orientation. Because of its primary interest, coin recall in Expt 1 always preceded stamp recall. However, in an experiment with participants unselected as to handedness, Richardson (1992) found that the incidence of correct left responses in stamp recall was greater when stamp testing preceded coin recall than when the order was reversed (though, in contrast, the level of coin recall was not significantly affected by order of testing). Thus in the following experiment the order of testing was balanced over coin and stamp recall. zyxwv 614 zyxwvuts zyxw G . V .Jones and Mavanne Martin EXPERIMENT 2 This experiment resembled Expt 1 except that the number of handedness categories and the number of testing orders were increased, and accordingly the number of participants was increased as well. Method Partdcipants zyx zyx zy There were 394 participants. On the basis of their reported hand preference in drawing (see Procedure), 186 were left-handed and 208 were right-handed. They were students or conference attenders at Oxford University, Oxford Brookes University, Warwick University or Worcester College of Higher Education, with an average age of 25.9 years (SD 9.7). The proportion of females was 49.7 per cent overall and did not differ significantly between left-handers and right-handers (~'(1,N = 394) = 0.09). Procedure Each participant was tested using a question booklet, with successive items printed on separate pages. For approximately half of the participants (193 participants) memory was probed first concerning coins and second concerning stamps, using the same questions as in Expt 1, while for the remainder (201 participants) the order was reversed. In either case, the two probes were followed by the same question concerning handedness for drawing as in Expt 1. Participants then completed Annett's handedness questionnaire. Results In the case of recall of coin head orientation, one participant did not produce a valid response, leaving 393 for analysis. There was no significant effect of order of testing (i.e. whether tested first or second) conditionalized upon handedness (i.e. reported hand preference in drawing) (~'(2,N = 393) = 2.92). Further analyses were therefore carried out on data aggregated over the two orders of testing. The proportion of left-handers who correctly recalled coin head orientation as facing to the right was 44.9 per cent (83/185), which did not W e r significantly from the chance level of 50 per cent (~'(1,N = 185) = 1.95). The same proportion for right-handers was 34.1 per cent (71/208), which was significantly below the 50 per cent level (~'(1,N = 208) = 20.94, p < .OOl). Most importantly, the proportion correct for left-handers was significantly greater than that for right-handers (1, N = 393) = 4.73, p < .05). The significant advantage of left-handed over right-handed participants in the recall of coin head orientation was examined further to determine if it held both for participants exhibiting pure handedness and those exhibiting mixed handedness. All 386 participants who completed the Annett questionnaire and did not respond ' either ' for writing hand preference were included. For pure left-handers, the proportion who correctly recalled head orientation as facing to the right was 46.9 per cent (38/81), which did not differ significantly from the chance level of 50 per cent ( ~ ~ (N1 = , 81) = 0.31). The same proportion for pure right-handers was 32.2 per cent (49/152), which was significantly below the 50 per cent level (~'(1,N = 152) = 19.18, p < .OOl). Most importantly, the proportion zyxwvu (x' zyxw zyxwv zyxwvu zyxwv z Handedness dependency .in recall 615 correct for pure left-handers was significantly greater than that for pure right-handers (~'(1,N = 233) = 4.86, p < .05). For mixed left-handers, the proportion who correctly recalled head orientation as facing to the right was 42.9 per cent (42/98), which did not differ significantly from , 98) = 2.00). The same proportion for the chance level of 50 per cent ( ~ ~ (N1 = mixed right-handers was 38.2 per cent (21/55), which also did not differ significantly from the 50 per cent level ( ~ ~ (N1 = , 55) = 3.07). Most importantly, the proportion correct for mixed left-handers did not differ significantly from that for mixed righthanders (x2(1,N= 153) = 0.32). The relation between degree of handedness and recall of coin head orientation was also examined by regression analysis. Recall score was 0 and 1 for left (incorrect) and right (correct) responses, respectively, and the handedness index ranged from 12 (complete right) to 36 (complete left). There was a significant linear regression of recall score on handedness index (F(1,387) = 6.79, p < .Ol), but no significant quadratic or higher trend. If p and a represent recall and handedness, respectively, the best-fitting equation was p = 0.00673a 0.242. Turning to the recall of stamp head orientation and reverting to handedness in terms of reported hand preference in drawing, there was a significant effect of order of testing conditionalized upon handedness (~'(2, N = 394) = 16.27,p < .OOl). The effect was confined to right-handers (~'(1,N = 208) = 16.19, p < .OOl), with no significant effect of order upon left-handers ( ~ ~ (N1 = , 186) = 0.08). Further analyses were therefore carried out on data aggregated over the two orders of testing for lefthanders, but on unaggregated data for right-handers. The proportions of left-handers, of right-handers tested first and right-handers tested second who correctly recalled stamp head orientation as facing to the left were 68.3 (127/186), 83.5 (86/103) and 58.1 per cent (61/105), respectively. The first two proportions were significantly above the 50 per cent level but the third one was not (x2(1,N= 186) = 24.86, p < .001; x'(1,N = 103) = 46.22, p < .001; and (1, N = 105) = 2.75, respectively). There was a significant difference among the three , 394) = 16.10,~< .OOl). However, the level of correct stamp proportions ( ~ ~ (N2 = recall by left-handers (i.e. 68.3 per cent) did not differ significantly from that for right-handers overall (i.e. 70.7 per cent) (x'(1,N = 394) = 0.26), and thus was not examined further with respect to finer-grained measures of handedness. + x' Discussion The results of Expt 2 confirmed and extended those of Expt 1 in several ways. First, they confirmed all the principal findings. There was a reliable difference between lefthanded and right-handed people (with handedness defined in terms of the hand preferred for drawing) in their recall from everyday memory. Left-handed people were again shown to be significantly more likely than right-handed people to remember correctly that the Queen's head on a British coin faces to the right. Similarly, it was again found to be only the right-handed population for which a significant majority misremembered the direction in which the coin head faces. There was in contrast again no significant difference between left-handers and right-handers in stamp head recall, with the magnitude of the left-handedness advantage zyxwvutsr zyxwvutsrq zyx zyxw 616 G. V.Jones and Maryanne Martin numerically smaller in this experiment both for coin recall, 10.8 per cent, and stamp recall, 2.4 per cent. Second, new types of analysis were carried out on coin recall performance using participants’ responses to the questionnaire of Annett (1970, 1985). These showed that the overall difference in recall patterns between left-handers and right-handers derived from only some of the participants. Pure left-handers recalled significantly better than pure right-handers, but mixed left-handers and mixed right-handers performed at an intermediate level and did not differ significantly from each other. Alternatively analysed, there was a significant linear relation between correct recall and degree of left-handedness. Bryden, McManus & Bulman-Fleming (1994b, p. 315) have pointed out that in principle the use of any classification of handedness more complex than a simple dichotomy between left-handers and right-handers may make the achievement of comparability across studies more difficult. However, the two more complex forms of classification used here yielded highly consistent conclusions. Finally, concerning the order of testing, the principal result was that there was no significant effect of the order of testing of coin recall for left-handers or for righthanders. For stamp recall, there was again no significant effect of order of testing for left-handers, but the incidence of correct left responses by right-handed participants was greater when stamp testing preceded coin testing than when the order was reversed. Richardson (1992) tested coin and stamp recall with a group of participants unselected as to handedness, and thus mainly right-handed. He found the same pattern of results, with no significant effect of order on coin recall but significantly higher correct recall of the stamp head when this was tested first. GENERAL DISCUSSION The results of the two experiments reported here demonstrate the existence of a reliable effect of handedness in a verbal test of everyday memory. When asked in which direction the Queen’s head faces on a British coin, significantly more righthanded people than left-handed people recalled incorrectly that the head faces to the left. Indeed, right-handed people exhibited systematic misremembering in that for them (but not for left-handed people) overall performance was significantly worse even than the 50 per cent level to be expected by chance. Previous observations of systematic misremembering in British coin head recall (e.g. Jones, 1990; Jones & Martin, 1992; Richardson, 1992) have explained the phenomenon in terms of the schema model of Rubin & Kontis (1983). A single schema is formed to represent the head appearing on coins (right-facing) and on stamps (left-facing) and tends predominantly to reflect the latter because of the greater frequency of viewing it appropriately oriented (on an envelope, rather than in a handful of change). However, the present results and those of McKelvie & Aikins (1993) suggest that a second mechanism, related to handedness, is also operative. This inference is consistent also with the results of a recent study of memory for the Danish coinage (Martin & Jones, 1995). Like British coins, Danish coins bear the profile of a Queen’s head (in this case, Queen Margrethe 11) facing to the right but, unlike the British case, this information is not counteracted by a leftfacing profile on stamps. Despite the absence of a misleading schema, however, zyxw zyxw zyxwvut Handedness dependency in recall 617 significant misremembering of the coin profile was found also in Denmark among a sample not selected on the basis of handedness and hence presumably predominantly right-handed. It is thus proposed that there are two separate factors which increase the likelihood that a person suffers from the mnemonic illusion that a right-facing head on a coin faces to the left. The first is that the person’s relevant schema contains incorrect information (e.g. acquired from exposure to British postage stamps). The second is associated with right-handedness. What is the origin of the handedness effect in the recall paradigm under investigation? It is proposed here that it derives from asymmetrical motor imagery (as opposed to the static imagery investigated by McKelvie & Aikins, 1993, Expt 3). Motor imagery refers to a form of mental representation which is related to motor physiology in the same way as visual imagery is related to visual physiology. ‘To illustrate, consider a pupil learning a motor skill such as playing a musical instrument. ...he must form an image in his mind of the teacher’s action. Conversely, when the teacher watches the pupil’s repetition, though not performing the action himself, he will experience a strong feeling of what should be done and how. ... The teacher’s feelings during the pupil’s performance would be based on the discharge of the same neurons in his own brain that were firing while he was preparing and executing the correct movements. According to this hypothesis, these “representation ’’ neurons are the same as those activated during preparation for actual action’ (Jeannerod, 1994, p. 189). On the motor imagery view, then, even if a person is asked to recall verbally the direction in which a head faces on a coin there may be a tendency to activate relevant motor representations. If the geometry of physical hand movements favours the drawing of left-facing profiles by right-handed people, then the motor imagery which is activated will reflect this asymmetry and will predispose memory performance in the direction of a left response by right-handed people, and vice versa, as observed. Is there any evidence that right-handed people find it easier to draw left-facing profiles, and vice versa, as the motor imagery explanation of the present findings requires? Shanon (1979) instructed participants simply to draw a face in profile, and observed precisely this pattern of performance. The majority of right-handed people, 81.3 per cent (65/80), drew a left-facing profile, but only a minority of left-handed people, 43.7 per cent (35/80), did so; the difference is highly significant (~‘(1,N = 160) = 24.00). Richardson (1992) obtained a difference in the same direction which did not reach significance, possibly because he had only a small left-handed group: 64.1 per cent (82/128) of right-handers and 57.9 per cent (11/19) of left-handers drew a left-facing profile. The evidence thus is that there is indeed an underlying asymmetry in manual behaviour regarding the depiction of head profiles which may, via motor imagery, be reflected in cognitive performance. With respect to everyday memory, do left-handers and right-handers each form a uniform group? The evidence from Expt 2 was that they do not. Rather, it was found that as handedness changed by gradations from pure right-handed to pure lefthanded, so coin recall performance improved by degree. It may be noted that such a finding raises the possibility of helping to elucidate the origins of handedness itself-for example, by discriminating among alternative genetic models of handedness (e.g. Annett, 1985; McManus, 1985; McManus & Bryden, 1993). z zyxwvutsr zyxwvutsr zyxwv zyxw 618 G . V.Jones and Maryanne Martin Finally, it is perhaps worth noting that the present view suggests that future research may uncover other aspects of cognitive performance which differ between left-handed and right-handed people. On the present account, handedness effects in cognition are in principle possible whenever cognitive performance is influenced by motor imagery which takes two distinct forms for left-handed and for right-handed manual behaviour, even if no actual manual behaviour is required. At present, evidence of at least one such effect of handedness upon recall from everyday memory does appear to be secure. zyxwv zy Acknowledgements The authors are grateful to Jennifer Bauwens, Angela Emler and Claire Hawkins for help with testing, and to Stuart McKelvie, Chris McManus and two anonymous reviewers for helpful comments. The first author is grateful to St Cross College, University of Oxford, for the facilities offered by a Visiting Fellowship. References Annett, M. (1970). A classificationof hand preference by association analysis. BritishJournal of Psychology, 61, 303-321. Annett, M. (1985). Left, Right, Hand and Brain: The Right Shqt Theory. London: Erlbaum. Bishop, D. V. M. (1984). Using non-preferred hand skill to investigate pathological handedness in an unseleaed population. Developmental Medicine and Child Neurology, 26, 214-226. Bishop, D. V. M. (1990). Handedness and Developmental Disorder. Oxford : Blackwell. Bruce, V. (1989). Human factors in the design of coins: 1989 C. S. Myers lecture. The Psychologist, 12, 524-527. Bryden, M. P., McManus, I. C. & Bulman-Fleming, M. B. (1994~).Evaluating the empirical support for the Geschwind-Behan-Galaburda model of cerebral lateralization. Brain and Cognition, 26, 103-1 67. Bryden, M. P., McManus, I. C. & Bulman-Fleming, M. B. (1994b). GBG, BMB, R & L, X & Y ...Reply to commentaries. Brain and Cognition, 26, 312-326. Eglinton, E. & Annett, M. (1994). Handedness and dyslexia: A meta-analysis. Perceptual and Motor Skills, 79, 1611-1616. Geschwind, N. & Behan, P. (1982). Left-handedness : Association with immune disease, migraine, and developmental learning disorder. Proceedings of the National Academy of Sciences of the U S A , 79, 5097-5100. Geschwind, N. & Galaburda, N. (1987). Cerebral Lateraliqation: Biological Mechanisms, Associations and Pathology. Cambridge, MA: MIT Press. Jeannerod, M. (1994). The representing brain: Neural correlates of motor intention and imagery. Behavioral and Brain Sciences, 17, 187-245. Jones, G. V. (1990). Misremembering a common object: When left is not right. Memory and Cognition, 18, 174-182. Jones, G. V. & Martin, M. (1992). Misrememberinga familiar object: Mnemonic illusion, not drawing bias. Memov and Cognition, 20, 211-213. McKelvie, S. J. & Aikins, S. (1993). Why is coin head orientation misremembered? Tests of schema interference and handedness hypotheses. British Journal of Psychology, 84, 355-363. McManus, I. C. (1985). Handedness, language dominance and aphasia: A genetic model. Psychological Medicine, Monograph Supplement 8. McManus, I. C. & Bryden, M. P. (1993). The neurobiology of handedness, language and cerebral dominance: A model for the molecular genetics of behaviour. In M. H. Johnson (Ed.), Brain Development and Cognition: A Reader. Oxford : Blackwell. Marshall, J. C. (1981). Lateral and focal organization in the human brain. In Y. Lebrun & 0. Zangwill (Eds), Lateralisation of Lnguage in the Child, 71-81. Lisse, The Netherlands: Swets & Zeitlinger. Martin, M. & Jones, G. V. (1992). Exploring a mnemonic illusion: When does General de Gaulle turn to the left? Cahiers de Psychologie Cognitive, 12, 103-113. zyxw zy zyxwvutsr zyxwvut Handedness dependency in recall 619 zy zy Martin, M. & Jones, G. V. (1995). Danegeld remembered: Taxing further the coin head illusion. Memory, 3, 97-104. Nickerson, R. S. & Adams, M. J. (1979). Long-term memory for a common object. Cognitive Psychology, 11, 287-307. Orton, S . T. (1925). ‘Word-blindness’ in school children. Archives of Neurology and Psychiatry, 14, 581-615. Richardson, J. T. E. (1992). Remembering the appearance of familiar objects: A study of monarchic memory. Bulletin of the Psychonomic Society, 30, 389-392. Richardson, J. T. E. (1993). The curious case of coins: Remembering the appearance of familiar objects. The Psychologist, 6, 36&366. Rosenthal, R. (1991). Meta-analytic Procedures for Social Research. London : Sage. Rubin, D. C. & Kontis, T. C. (1983). A schema for common cents. Memory and Cognition, 11, 335-341. Shanon, B. (1979). Graphological patterns as a function of handedness and culture. Neuropsychologia, 17, 457-465. Received 25 September 199s; revised version received 23 Ju& 1996