ࡱ> 5@ bjbj22 'XX?RRRRR&N Znz'z'z'8'4(\n7N)d)@)));+;+;+c6e6e6e6e6e6e6$8R0;6,i17+;+i1i16RR))7|333i1Rl))c63i1c634314r5T,6)B) Їhtz'2 5G67`75;3;6nnRRRR;<6@;+V,3-I. ;+;+;+66nD6&D3pn6&The context-sensitivity of visual size perception varies across cultures M. Doherty a*, H. Tsuji b, and W. A. Phillips a, c a Department of Psychology, ҪԹ, Stirling FK9 4LA, UK b Department of Psychology, Faculty of Human Science, Osaka Shoin Women's University, 958 Sekiya, Kashiba, Nara, 639-0298 Japan c Frankfurt Institute for Advanced Studies, Johann Wolfgang Goethe-Universitat, Max-von-Laue-Strae 1, 60438 Frankfurt am Main, Germany * Correspondence to: m.j.doherty@stir.ac.uk Submitted First draft submitted 7th November 2007 Revision submitted 4th April 2008 The context-sensitivity of visual size perception varies across cultures Abstract. There is evidence that East Asian cultures have more context-sensitive styles of reasoning, memory, attention, and scene perception than western cultures. Lower levels of the perceptual hierarchy seem likely to be more similar in allhomogeneous across cultures, however, so we compared context-sensitivity in Japan with that in the UK using a rigorous psychophysical measure of the effects of centre-surround contrast on size discrimination. In both cultures context-sensitivity was greater for females working in the social sciences than for males working in the mathematical sciences. More surprisingly, context-sensitivity was also much greater in Japan than in the UK. These findings show that, even at low levels of the visual processing hierarchy, context-sensitivity varies acrossdepends upon cultures, and they raise important issues for both vision scientists and cross-cultural psychologists. 1 Introduction Context-sensitivity is central to cognition in general, and to perception in particular. Within the visual system much is known about its anatomical and physiological bases (Gilbert 1995; Phillips and Singer 1997; Wrgtter and Eysel 2000), and about its role in psychopathology (Hemsley 2005; Phillips and Silverstein 2003; Uhlhaas and Silverstein 2005). Contextual interactions can be defined as those that affect signal processing without changing what the signals mean (Kay et al 1998; Phillips and Craven 2000; Smyth et al 1996). Contextual interactions are neither necessary nor sufficient for signal transmission, but their modulatory effects play a major role in producing a coherent response to the situation as a whole. TOn the basis of many studies of the effects of context two broad cognitive styles have been distinguished: one that has low context-sensitivity and is therefore local, field-independent, and focussed on detail, and one that has high context-sensitivity and is therefore global, field-dependent, and focussed on gist (e.g. Happ and Frith 2006). Males tend to have low context-sensitivity (Phillips et al 2004), as do people with autistic spectrum disorders (weak central coherence; Frith 2003). Baron-Cohen (2002) argues that autistic spectrum conditions are extreme forms of this male style. Independently of these advances within cognitive neuroscience and neuropsychiatry, there has been a growing interest in cross-cultural comparisons between cognitive styles in East Asian and western cultures. Within anthropology the concept of high context and low context societies has arisen to distinguish between highly communal and highly individualistic societies (Hall 1976). Within cross-cultural psychology the distinctions between collectivistic and individualistic cultures, and between holistic and analytic cognitive styles now plays a central rolehas been shown to be important (Nisbett et al 2001; Nisbett and Masuda 2003; Heine, 2008). Nisbett and Miyamoto (2005) conclude that people in individualistic western cultures tend to engage in context-independent analytic perceptual processes, whereas people in East Asian cultures, such as China, Korea, and Japan, tend to engage in context-dependent holistic perceptual processes. Most of the evidence for these cultural differences in context-sensitivity comes from studies of high-level processes, such as the attribution of causality, categorization, attention, working memory, and scene perception. Cultural differences now need to be studied using the rigorous psychophysical methodologies that have recently been developed to study context-sensitivity in low-level vision (e.g. Kvacs 1996). These include the effects of context oncentre-surround contrasts of brightness, colour, and size perception (e.g. Phillips et al 2004), flanker effects on edge-detection (e.g. Polat and Sagi 1993), and contour integration (e.g. Field et al 1993),. These methodologies have been used intensively to study low-level contextual interactions in vision, but cross-cultural studies using these methods are rare. As the underlying processes occur automatically, and cannot be over-ridden by voluntary attentional strategies, the prevailing assumption is that these basic low-level processes are common to all cultures. The results reported here suggest that this view may be qualitatively correct, but quantitatively incorrect, i.e., low-level contextual interactions occur in all cultures, but to different extents in different cultures. Kitayama et al (2003) have shown that reproduction of line-length is more context-sensitive in Japanese than in western cultures. Their task involved inspecting a line drawn within a frame, and drawing a line of the same absolute or relative length within another frame of either the same or a different size.and then drawing it within another frame of either the same or a different size. To ensure that low-level perceptual processes, such as iconic memory, played no role the second frame was placed in the opposite corner of the laboratory. The cultural differences that they observed are therefore more likely to have arisen from differences in attention, working memory, or motor control than from any differences in low-level vision. A recent study of the Himba, a remote semi-nomadic culture in northern Namibia, suggests that contextual interactions in low-level vision may vary across cultures. They showed significantly less context-sensitivity than people in the UK, as measured by the Ebbinghaus illusion (de Fockert et al 2007). Though suggestive, this study is not conclusive, because group sizes were small (8 in each culture), matching subjects across such widely differing cultures is difficult, and there is no evidence concerning the extent of context sensitivity in the Himba at higher cognitive levels. In the experiments reported here weWe therefore studied differences between people in Japan and in the UK using a rigorous psychophysical paradigm that specifically measures the effect of surrounding context on visual size discriminationbased on the Ebbinghaus illusion (Phillips et al 2004), as shown in . This paradigm is a recently developed version of the Ebbinghaus illusion, which was discovered more than a century ago, and about which much is now known (see Figs. 1 and 2). This studies allows the study of the effect of surrounding context on size perception. IIt has long been thought that such effects areize perception is predominantly pre-attentive (e.g. Stuart et al 1993), and that such simultaneous contrast effects involve activity at low levels of the visual pathways. Neuroimaging now provides further support for this view. Using functional MRI, Murray et al (2006) found that activity in the human primary visual cortex (V1) corresponds to the perceived size of objects as influenced by contextual cues, rather than to the retinal size of the objects image. The effect of context on size discrimination is a good candidate for cross-cultural studies study because, even though it is pre-attentive, it is not fully developed by four years of age (Kldy and Kovcs 2003), so there is plenty of time for socialisation to exert an influence. Furthermore, there is evidence that this effect can vary across cultures as : a recent study of the Himba, a remote semi-nomadic culture in northern Namibia, found that the Himba, a remote semi-nomadic culture in northern Namibia, they showed less context-sensitivity than people in the UK (de Fockert et al 2007). There is also evidence that different cognitive styles are associated with chosen profession. Fathers of autistic boys are more likely to be employed in 'detail focused' professions such as computing, accountancy, and engineering (Briskman et al 2001), and students of sciences in which mathematics is important, such as physics and engineering, report higher autistic spectrum traits than students of the humanities or social sciences (Baron-Cohen et al 2001). These differences have mostly been studied in western cultures, however, so to determine whether there are similar differences in Japan we compared females working in the social sciences with males working in the mathematical sciences. The success of Japanese engineers in global competition raises the possibility that they may be even more analytic, or detail-focussed, than their competitors in the west. _______________________ Figures 1 and 2 around here _______________________ 2 Method WThe paradigm that we used was developed through prior studies that produced a measure specific to the effects of context, sensitive to individual differences, easy for participants to understand, and requiring only a few minutes (Phillips et al 2004). It uses a two-alternative forced choice method paradigm (Phillips et al 2004) in which the task is to decide whether the centre circle that looks larger is in the 3 x 3 array on the left or in that on the right, and to indicate this by pressing the left-pointing or the right-pointing arrow on the keyboard. The two centre circles always differed in actual size, and this difference varied in magnitude across trials. Unlike most previous studies, our paradigm includes conditions in which centre-surround contrast enhances size discrimination. Without these conditions subjects participants could always choose the correct responserespond correctly simply by selecting the array with larger surrounds, so they are needed to unconfound differences in centre size from differences in surround size. 2.1 Participants In the main experiment, two groups of participants were tested in Japan: females in the social sciences and males in the mathematical sciences. These were 29 female students from the Department of Psychology at Osaka ShMin Women s University, and 34 male engineers. The engineers were 16 professional software or electrical engineers, mostly employed by Toshiba, and 18 mechanical engineering postgraduate students, working in the fluid mechanics laboratory at the KyMto Institute of Technology. Their performance was compared with that of staff and students working in the University of Stirling in the UK. Of these 16 were females from the Department of Applied Social Science, and 16 were males from the Department of Computing Science and Mathematics. Performance of the two UK groups was included in the report by Phillips et al (2004) where it was shown that there are independent effects of sex and profession. We co-varied them here to obtain cross-cultural comparisons at widely separated points on the within culture spectrum of sensitivity to context. Following the main study, two slightly modified experiments were run using additional female psychology students in Japan and in the UK. 2.2 Stimuli and experimental design On each trial two 3x3 arrays of circles were presented, side by side on a computer screen. The centre circle of one array was 100 pixels in diameter; the centre circle of the other was 2, 6, 10, 14, or 18 pixels larger or smaller. Each of these 10 size differences was presented 8 times, with the larger central circle surrounded by larger circles (125 pixels diameter) and the smaller central circle surrounded by smaller circles (50 pixels diameter). In these conditions size contrast impairs discrimination. If only these conditions were used participants could be correct on every trial simply by choosing the array with larger surrounds, which is easy to see. Therefore, to unconfound relations between surround sizes and centre sizes the 98 and 102 pixel circles were presented eight times each with the smaller centre circle surrounded by circles 125 pixels in diameter and the larger central circle surrounded by circles 50 pixels in diameter. Size contrast then increases accuracy if participants are judging the apparent sizes of the centre circles, but if they choose the array with larger surrounds then they will be wrong on every trial in this condition. The 96 trials [(10 x 8) + (2 x 8)] were presented in random order. 2.3 Apparatus A java Java program developed by Phillips et al (2004) was used to present stimuli and record and analyse responses. Stimuli were presented on laptop computers (in the UK on a Toshiba 4090 CDS laptop with a 13-inch screen; in Japan on either a Toshiba Dynabook SS M10 11L/2 with a 12-inch screen or a Toshiba Dynabook SS 1620 12L/2 with a 13-inch screen). Screens were set at 800 x 600 pixels resolution. 2.4 Procedure Participants were tested individually in a quiet room. They were shown an example of the stimulus arrays to be used and the task was explained. They were asked to use the cursor keys to indicate which central circle looked larger, irrespective of actual size. They were told that the circles were never the same size, and to guess if they were uncertain. Stimuli were presented for 2000 mseconds and subsequent trials followed roughly 2000 secondsms after response. No feedback was given during the 7 minute procedure. The distance between the participant and screen was that normal for laptop operation. 3 Results We first compare discrimination in the two conditions in which there was a 2% difference in size. In one of these, context was expected to make the size discrimination easier as shown in Fig. 1. In the other, it was expected to make the discrimination harder, as shown in Fig. 2. Only the context distinguishes these conditions so we can be confident that any effects are due to context alone. With helpful context, for Japanese male mathematical scientists discrimination accuracy was 99.8%; with misleading context, it was 3.3%, a highly significant difference (t28 = 50.6, p < 0.001, d = 16.30). For Japanese female social scientists discrimination accuracy was 96.8% with helpful context and 4.6% with misleading context (t33 = 66.9, p < 0.001, d = 17.59). With helpful context, for British male mathematical scientists discrimination accuracy was 97.3%; with misleading context it was 10.2% (t15 = 26.0, p < 0.001, d = 10.05). For British female social scientists discrimination accuracy was 96.5% with helpful context and 4.3% with misleading context (t15 = 38.6, p < 0.001, d = 16.71). The large effects of context when the size difference was only 2% show that all subjects were sensitive to context. How sensitive they were is shown by the real size difference needed to overcome the misleading effects of size contrast. The relation between accuracy and the actual size difference for the conditions where context impairs discrimination is shown in Fig 3. Accuracy was low when the real size difference was 2% but much higher when it was 18%, where most, but not all, of the misleading effects of context have been overcome. Clearly, a larger real size difference is needed to overcome the effects of context in Japan than in the UK. _______________________ Figure 3 around here _______________________ Overall context-sensitivity can be quantified by the total number of correct responses out of the 80 trials, incorporating all 5 size differences, for the conditions in which context impairs discrimination. This measure increases as context-sensitivity decreases. The distributions of individual scores on this measure are shown in Fig. 4. Though there were clear individual differences within groups and some large overlaps between groups, for some group comparisons there was little overlap. In Japan, 25 out of 29 female social scientists scored less than 40 correct out of 80 overall, compared with only 1 out of 16 UK male mathematical scientists. _______________________ Figure 4 around here _______________________ A 2-way ANOVA of total score showed a main effect for culture, F (1, 3) = 27.2, p < 0.001, p2 = 0.23, and a main effect for sex/discipline, F (1, 3) = 11.3, p = 0.001, p2 = 0.11. There was no interaction, F (1, 3) = 0.045, p = 0.83, p2 = 0.00. Planned t-tests showed a significant difference between the 2 cultural groups: t (93) = 4.842, p < 0.001, d = 1.08. There were also significant differences within each culture between the male mathematical scientist and female social scientist groups: for the two British groups, t (30) = 2.59, p = 0.015, d = 0.92; for the two Japanese groups, t (61) = 2.53, p = 0.014, d = 0.65. The only non-significant comparison between individual groups was between British female social scientists and Japanese male mathematical scientists, t (48) = 1.22, p = 0.228, d = 0.39. The above se results do not tell us how participantswhether discrimination in Japan rises to an asymptote similar to that in the UK with further increase in perform with real size differences greater than 18%. It may be that there is some difference between cultures that produces lower accuracy in Japan at all real size differences. To explore this we gave a new set of 16 Japanese female social science students the same task, but using real differences up to 28%, in steps of 6%. Discrimination rose to an asymptote level just below 100% at size differences of about 22%, as compared to about 14% in the UK. The rate at which discrimination rises to this asymptotelevel, which primarily reflects response reliability, seems to be much the same in the two cultures. What differs is not response reliability but the amount of real size difference required to overcome the misleading effects of context. In the above experiments participants were asked to say which of the two centre circles looked larger, irrespective of their actual size. It is possible that interpretation of this instruction varies across cultures. To explore this we tested new participants in both Japan and the UK. In Japan 25 further female psychology students were tested in Osaka ShMin Women s University. The task was exactly as before, except that the instructions were modified to encourage participants to try to  see through the illusion. They were told that the stimuli were examples of a well-known illusion, and that they were to judge which of the central circles was really larger irrespective of which looked larger. Twenty female psychology students studying at Stirling University in the UK were also tested: 10 were asked to select the centre circle that looked larger; the other 10 were asked to select that which was really larger, just as in Japan. The results show that for both cultures there were large effects of context even when subjects were warned about its illusory effects. In Japan the mean score out of 80 across all conditions where context was misleading was 23.28 3 for the looks larger instruction, and 29.768 for the really is larger instruction. In the UK the mean scores for these two instruction conditions were 30.6 and 40.4, respectively. This contrasts with scores of more than 95% correct for all groups when context is helpful. A 2-way ANOVA of these scores out of 80 was performed to compare the new results from the UK with those from Japan. There was a main effect for culture, F (1, 3) = 7.263, p = 0.009, p2 = 0.094, and a main effect of instruction condition, F (1, 3) = 5.857, p = 0.018, p2 = 0.077. There was no interaction, F = 0.257, p = 0.614, p2 = 0.004. These results show the effect of culture to be robust. They indicate that the cultural differences are not due to different interpretations of the instructions, and they provide further evidence that participants cannot voluntarily ignore the context. 4 General discussion These results are evidence that size perception is more context-sensitive in Japan than in the UK. They extend the evidence for cultural differences at higher levels of cognition and perception (e.g. Nisbett and Masuda 2003; Nisbett and Miyamoto 2005) to specifically include contextual interactions that occur at lower levels in the visual processing hierarchy. Nisbett et al (2001) argue that cognition may be far less universal across cultures than is often assumed. Our results show the importance of both commonalities and differences across cultures. The contextual interactions producing size contrast occurred in every individual tested in both cultures. Nevertheless, there was great variation in the strength of those interactions. Thus, the cultural difference does not involve the use of context in some all-or-none fashion. At higher cognitive levels it might be possible for there to be context-independent cultures, as suggested by Nisbett and Miyamoto (2005), but at the level of size contrast all cultures are context-dependent. In some cultures this dependence is large; in others it is very large. Furthermore, as we found differences across sex and profession within each culture, this emphasises the importance of careful participant matching in cross-cultural research. On present data we cannot rule out the possibility of a genetic influence on the demonstrated cross-cultural differences. However, we are inclined to explain the differences in terms of culture-specific patterns of attention. Nisbett and Masuda (2003) suggest that cultural differences in perception derive predominantly from differences in attention. If these differences existed simply at the level of moment-to-moment attentional focus, they could not explain our findings. In our size perception task all subjects attended to the relative size of the centre circles, and used that to determine response. Otherwise discrimination could not have been near 100% on trials where the surrounding context made the real size difference seem larger than it was. However, context must also have had a large effect on all trials, because discrimination was near 0% for the same real size difference when context made the relative sizes seem to be the opposite of what they really were. Thus on all trials both target and context influenced response. No observer could choose to ignore relations between centre and surround circles, even though they all attended to and judged the relative sizes of the centre circles. Furthermore, modifying the instructions so as to encourage participants to ignore the surrounds had the same effect in both cultures. Thus, cultural differences in this paradigm do not arise from short-term differences in patterns of attention. Nevertheless, it is likely that there are attentional effects in this task (Shulman, 1992), and furthermore there is extensive evidence for attentional effects throughout all levels of the visual system. We suggest that chronic differences in the pattern of attention between cultures may affect low-level vision through activity-dependent synaptic plasticity. We now need to discover the origins of these cultural differences in low-level vision. Differences in cultural practices and in their physical environments both make large contributions to differences in context-sensitivity at the higher levels of perception and cognition (Miyamoto et al 2006). The present study suggests that they may also influence low-level processes. 4 General discussion These results are evidence that size perception is more context-sensitive in Japan than in the UK. They extend the evidence for cultural differences at higher levels of cognition and perception (e.g. Nisbett and Masuda 2003; Nisbett and Miyamoto 2005) to specifically include contextual interactions that occur at lower levels in the visual processing hierarchy. Nisbett and Masuda (2003) conclude that cultural differences in perception derive predominantly from differences in attention. In our size perception task all subjects attended to the relative size of the centre circles, and used that to determine response. If not then discrimination could not have been near 100% on trials where the surrounding context made the real size difference seem larger than it was. However, context must also have had a large effect on all trials, because discrimination was near 0% for the same real size difference when context made the relative sizes seem to be the opposite of what they really were. Thus on all trials both target and context influenced response. No observer could choose to ignore relations between centre and surround circles, even though they all attended to and judged the relative sizes of the centre circles. Furthermore, modifying the instructions so as to encourage participants to ignore the surrounds had the same effect in both cultures. Thus, cultural differences in this paradigm do not arise from differences in the stimuli or relations that are selected to influence response. Nevertheless, there may still be some attentional effects in this task (Shulman 1992), and this is likely given the extensive evidence for attentional effects throughout all levels of the visual system. However, if there are, then they modulate contextual interactions that affect activity at low levels in the hierarchy of visual processing. Nisbett et al (2001) argue that cognition may be far less universal across cultures than is often assumed. Our results show the importance of both commonalities and differences across cultures. The contextual interactions producing size contrast occurred in every individual tested in both cultures. Nevertheless, there was great variation in the strength of those interactions. Thus, the cultural difference does not involve the use of context in some all-or-none fashion. At higher cognitive levels it might be possible for there to be context-independent cultures, as suggested by Nisbett and Miyamoto (2005), but at the level of size contrast all cultures are context-dependent. In some cultures this dependence is large; in others it is very large. . Japanese engineers have a perceptual style that is about as context-sensitive as that of females working in the social sciences in the UK. Thus, a highly context-sensitive, or analytic, perceptual style seems not to be necessary for success in engineering professions. This raises the possibility that Japanese success in engineering arises more from better social organization than from the individual cognitive style of those in the profession. Furthermore, as we found differences across sex and profession within each culture, this emphasises the importance of careful participant matching in cross-cultural research. We now need to discover the origins of these cultural differences in low-level vision. Differences in cultural practices and in their physical environments both make large contributions to differences in context-sensitivity at the higher levels of perception and cognition (Miyamoto et al 2006). thatmay also synaptic They may also influence low-level processess, especially ones that develop slowly through childhood, such as those studied here, possibly through effects on attention, which affect low-level visual processes, and which thus could affect low-level vision through activity-dependent plasticitysize contrast (Kldy and Kovcs 2003). Genetic contributions to the differences found here are also possible, and should be taken seriously as, in addition to the sex differences, it has been shown that the c Context-sensitivity studiedin the task used here ishas been shown to be greatly weakened in both disorganized schizophrenia (Uhlhaas and Silverstein 2005) and autism (Happ 1999). Strikingly, in both pathologies, performance in this task is supra-normal, rather than sub-normal, because the task is designed so that, atypically, context is misleading. It is not yet known whether a cultural bias toward high context-sensitivity is protective or detrimental in these disorders, so cross-cultural comparisons such as those made here may provide useful information on this issue. The findings reported here are limited in several ways. More professions could have been compared. Furthermore, we studied only one form of context-sensitivity in low-level vision. Others now need to be examined, including tasks where the use of context enhances rather than degrades performance. Here, we have also provided no cross-cultural developmental data for this form of context-sensitivity, nor any evidence of cultural differences in the underlying neuronal interactions. All of these limitations can be overcome by combining the rigorous methods of psychophysics and cognitive neuroscience with cross-cultural comparisons. The results reported above may encourage the further development of such a cross-cultural psychophysics, because although the basic underlying contextual interactions occurred in all cultures, their strength varied greatly both within and between cultures. Acknowledgements. We thank Jules Davidoff, Nick Humphrey, and Helen Ross for expert commentaries on an earlier version of this paper. We also thank the Departments of Psychology at Stirling University and the Osaka Shoin Womens University for supporting our research. References Baron-Cohen S, 2002 The extreme male brain theory of autism Trends in Cognitive Sciences 6 216 - 222 Baron-Cohen S, Wheelwright S, Skinner R, Martin J, Clubley E, 2001 The autism-spectrum quotient (AQ): Evidence from Asperger Syndrome/High-functioning autism, males and females, scientists and mathematicians Journal of Autism and Developmental Disorders 31 5 - 17 Briskman J, Happ F, Frith U, 2001 Exploring the cognitive phenotype of autism: Weak 'central coherence' in parents and siblings of children with autism: II. Real-life skills and preferences Journal of Child Psychology and Psychiatry 42 309 - 316 de Fockert J, Davidoff J, Fagot J, Parron C, Goldstein J, 2007 More accurate size contrast judgements in the Ebbinghaus illusion by a remote culture Journal of Experimental Psychology: Human Perception and performance 33 738 - 742 Field D J, Hayes A, Hess R F, 1993 Contour integration by the human visual system: Evidence for a local association field Vision Research 33 173 79 Frith U, 2003 Autism: Explaining the Enigma (Oxford: Blackwell)Gilbert C D, 1995 Dynamic properties of adult visual cortex In The Cognitive Neurosciences Ed. M S Gazzaniga (MA: MIT Press) pp 73 - 90 Hall E T, 1976 Beyond Culture (New York: Anchor) Happ F, 1999 Autism: cognitive deficit or cognitive style? Trends in Cognitive Sciences 3 248 - 254 Happ F, Frith U, 2006 The weak coherence account: Detail-focused cognitive style in autism spectrum disorders Journal of Autism and Developmental Disorders 36 5 - 25 Heine S J, 2008 Cultural Psychology (New York: Norton)Hemsley D R, 2005 The schizophrenic experience: Taken out of context? Schizophrenia Bulletin 31 1 - 11 Ji L, Peng K, Nisbett R E, 2000 Culture, control, and perception of relationships in the environment Journal of Personality and Social Psychology 78 943 - 955 Kldy Z, Kovcs I, 2003 Visual context integration is not fully developed in 4-year-old children Perception 32 657 - 666 Kay J, Floreano D, Phillips W A, 1998 Contextually guided unsupervised learning using multivariate local processors Neural Networks 11 117 - 140 Kitayama T, Duffy S, Kawamura T, Larsen J T, 2003 Perceiving an object and its context in different cultures Psychological Science 14 201 - 206 Kovcs I, 1996 Gestalten of today: Early processing of visual contours and surfaces Behavioral Brain Research 82 1 - 11 Miyamoto Y, Nisbett R E, Masuda T, 2006 Culture and the physical environment Psychological Science 17 113 - 119 Murray S O, Boyaci H, Kersten D, 2006 The representation of perceived angular size in human primary visual cortex Nature Neuroscience 9 429 - 434 Nisbett R E, Masuda T, 2003 Culture and point of view Proceedings of the National Academy of Sciences U. S. A. 100 11163 - 11175 Nisbett R E, Miyamoto Y, 2005 The influence of culture: holistic versus analytic perception Trends in Cognitive Sciences 9 467 - 473 Nisbett R E, Peng K, Choi I, Norenzayan A, 2001 Culture and systems of thought: Holistic versus analytic cognition Psychological Review 108 291 - 310 Phillips W A, Chapman K L S, Berry P D, 2004 Size perception is less context-sensitive in males Perception 33 79 - 86 Phillips W A, Craven B J, 2000 Interactions between coincident and orthogonal cues to texture boundaries Perception and Psychophysics 62 1019 - 1038 Phillips W A, Singer W, 1997 In search of common foundations for cortical computation Behavioral and Brain Sciences 20 657 - 722 Phillips W A, Silverstein S M, 2003 Convergence of biological and psychological perspectives on cognitive coordination in schizophrenia Behavioral and Brain Sciences 26 65 - 137 Polat U, Sagi D, 1993 Lateral interactions between spatial channels: Suppression and facilitation revealed by lateral masking experiments Vision Research 3 993 - 999 Shulman G L, 1992 Attentional modulation of size contrast Quarterly Journal of Experimental Psychology: A 45 529 - 546 Smyth D, Phillips W A, Kay J, 1996 Measures for investigating the contextual modulation of information transmission Network: Computation in Neural Systems 7 307 - 316 Stuart G W, Bossomaier T R J, Johnson S, 1993 Preattentive processing of object size: implications for theories of size perception Perception 22 1175 - 1193 Uhlhaas P J, Phillips W A, Schenkel L, Silverstein S M, 2006 Theory of mind and perceptual context processing in schizophrenia Cognitive Neuropsychiatry 11 416 - 436 Uhlhaas P J, Silverstein S M, 2005 Perceptual organization in schizophrenia spectrum disorders: Empirical research and theoretical implications Psychological Bulletin 131 618 - 632 Wrgtter F, Eysel U F, 2000 Context, state and the receptive fields of striatal cortex cells Trends in the Neurosciences 23 497 503  Fig. 1. An example of the stimulus display used in conditions where size contrast enhances size discrimination. The centre circle on the right is 2% larger than that on the left. Here it is surrounded by smaller circles which make it look larger than it is, whereas that on the left is surrounded by larger circles, which make it look smaller than it is. It is therefore easy to see that the centre circle on the right is larger than that on the left.  Fig. 2.IKSTUV`abvwxy{|~D E F źӬźӞxj\J"hXhQ5CJH*aJmHsHhhQ6CJH*aJh#hQ5CJH*aJhXhQ6CJaJhhQ6CJaJhXhQ5CJH*aJhSdhQ5CJ H*aJ h7hQ5CJ H*aJ hQ5CJH*aJhhQ5CJH*aJh7hQ5CJH*aJhQ5CJ aJ hhQ5CJ aJ hQ5CJ(aJ(I|E HNdC$Eƀ]gdQ d]gdQMdEƀ]gdQ$d]a$gdQTF G      ' , - A C N Ǽ{o{dVH/H1Hhu$Ħh}fh}fCJaJu$Ħ*H*Hhu$Ħh}fCJaJHht$ĦhQCJaJhwhQCJaJhfA6hQCJH*aJhQCJaJ!Hht$Ħhwh}fCJaJHht$Ħh}fCJaJ3hwhQh}fCJaJcHdhdhdht$Ħh&hQCJaJhQ6CJaJmHsHhQ6CJaJmHsHhhQ6CJaJmHsHhXhQCJaJmHsH - O P -.02Aprd]`gdQ$d]a$gdQ$  d]a$gdQNdC$Eƀt$Ħ]gdQ d]gdQ N O Q u x z  sj^I^)hQh&CJcHdhdhdhĆHhĆhQCJhshQCJ)hQh&CJcHdhdhdhĆHhĆhQCJHhĆhQCJ)hQh&CJcHdhdhdhĆHh'ĆhQCJ hQCJhQ5CJaJhQCJaJhwhQ5CJ aJ hQ5CJ aJ hwhQCJaJh}fCJaJ -.125@A!*/01AY`gp$)ȿȴxaxax-hQh=nCJaJcHdhdhdhĆ3hYhQh=nCJaJcHdhdhdhĆhQCJaJ3hYhQh!CJaJcHdhdhdh"ĦhYhQCJaJhQ5CJaJh@`5hQ5CJaJhwhQCJaJ hQ5CJ hQ5CJ hQCJHhĆhQCJ)4>JPU,ST02CWY^dg pûí|ûnûnûûûûT3hYhQhBCJaJcHdhdhdhĆHhĆhQCJaJ-hQh&CJaJcHdhdhdhĆ3hYhQh&CJaJcHdhdhdhĆHhĆhQCJaJhQCJaJhYhQCJaJ-hQh=nCJaJcHdhdhdhĆ3hYhQh=nCJaJcHdhdhdhĆpq,=!'.+F+b+z+|+///}} d]gdQ$d]`a$gdQd]`gdQRdC$EƀĆ]`gdQpq#$,9;<=>ij樚fXJHhĆhQCJaJHh%#ĦhQCJaJ3hYhQh!CJaJcHdhdhdh"Ħ3hYhQh!CJaJcHdhdhdh"ĦHh"ĦhQCJaJHh"ĦhQCJaJHh"ĦhQCJaJhQCJaJ3hYhQhBCJaJcHdhdhdhĆhYhQCJaJHhĆhQCJaJ;K,5Ev  (-8Ժr[rMHhĆhQCJaJ-hQh75CJaJcHdhdhdhĆHhĆhQCJaJh~FhQCJaJHh"ĦhQCJaJHhĆhQCJaJhQCJaJhYhQCJaJ3hYhQh75CJaJcHdhdhdhĆHhĆhQCJaJ:HhĆhQh!CJaJcHdhdhdh"Ħ*+g2 pCQû軥|bH1H-hQhnFCJaJcHdhdhdhĆ3hYhQhnFCJaJcHdhdhdhĆ3h@VhQh@VCJaJcHdhdhdhcĆ.HhcĆh@VhQaJcĆ*CJ!HhcĆh@VhQCJaJh@VhQCJaJh+hQCJaJhQCJaJHhĆhQCJaJ-hQh75CJaJcHdhdhdhĆ-hQh75CJaJcHdhdhdhĆ ! " =!`!c!l!!!βtZC;0hYhQCJaJhQCJaJ-hQh75CJaJcHdhdhdhĆ3hYhQh75CJaJcHdhdhdhĆHhĆhQCJaJ-hQhnFCJaJcHdhdhdhĆ0hQhnF>*CJaJcHdhdhdhĆ6hYhQhnF>*CJaJcHdhdhdhĆ3hYhQhnFCJaJcHdhdhdhĆ-hQhnFCJaJcHdhdhdhĆ!!!!""""*"J"Z"`"n"o"p""" #########+#nn`B:HhĆhQh@VCJaJcHdhdhdhdĆHhĆhQCJaJ3hYhQh75CJaJcHdhdhdhĆ-hQh75CJaJcHdhdhdhĆHhĆhQCJaJhQCJaJhYhQCJaJHhĆhQCJaJ-hQh75CJaJcHdhdhdhĆ3hYhQh75CJaJcHdhdhdhĆ+#?#t#u#v#w#x########$$$!$F$N$]$a$$$$$U%u%y%%ջ㻰wi[PhuhQCJaJHhJ$ĦhUCJaJHhĆhQCJaJ-hQhnFCJaJcHdhdhdhĆ3hYhQhnFCJaJcHdhdhdhĆhQCJaJhYhQCJaJ3hYhQhnFCJaJcHdhdhdhĆHhĆhQCJaJHhĆhQCJaJHhdĆhQCJaJ%%%%&7&A&G&H&P&p&&&&&&&+'6'seT3e@HhĆhYhQh@VCJaJcHdhdhdhfĆ!HhĆhYhQCJaJHhĆhQCJaJHhfĆhQCJaJ:HhĆhQh@VCJaJcHdhdhdhfĆHhĆhQCJaJhuhQCJaJhQCJaJHheĆhQCJaJ3huhQh@VCJaJcHdhdhdheĆ-hQh@VCJaJcHdhdhdheĆ6'r't'u'y'''''''''''($(r(|(((e**y+z+|+¡o_TIAIAIAIAIAIAIhQCJaJhYhQCJaJhuhQCJaJHhĆhQ>*CJaJ$HhĆhYhQ>*CJaJ!HhĆhYhQCJaJHhĆhQCJaJ@HhĆhYhQh@VCJaJcHdhdhdhfĆ:HhĆhQh@VCJaJcHdhdhdhfĆHhfĆhQCJaJ!HhfĆhYhQCJaJ|+~+++++++.,:,j,t,z,,,,,,,,,·lUllG6G6!HhĆhYhQCJaJHhĆhQCJaJ-hQhnFCJaJcHdhdhdhĆ3hYhQhnFCJaJcHdhdhdhĆ-hQhnFCJaJcHdhdhdhĆ3hYhQhnFCJaJcHdhdhdhĆhYhQCJaJ3hYhQhnFCJaJcHdhdhdhĆHhĆhQCJaJhYhQ5CJaJhQ5CJaJ,,----.W.g...... //////////F0000:1x111233,3ĭĭēsfhQCJaJnHo(tHhYhQ6CJaJhQ6CJaJh@`5hQCJaJhYhQ5CJaJHhmĆhQCJaJ-hQh~CJaJcHdhdhdhmĆhQCJaJ3hYhQh~CJaJcHdhdhdhlĆhYhQCJaJHhĆhQCJaJ#,3@3F34444b6d6h668%9&919=9O9]9^9c9i9n9o999%:D::::!;d;h;r;t;y;z;~;<!<K<Q<S<V<[<^<<< ==== ===>黭hQCJaJnHo(tH!Hhe$ĦhYh =CJaJHhe$Ħh =CJaJ3hYhQh =CJaJcHdhdhdhe$ĦhYhQ6CJaJhQCJaJhYhQCJaJhQCJaJnHtH7/c6d6c;d;r; = =??C]FuFFFF2IJI_IwIxINQ\\$d]`a$gdQd]`gdQ d]gdQ>>>>>>>>>>>>>?????AAbBdB CCCCCCIDZD4G5GwIxIJJJJ̾̾ڶڕڕڕڕډڶoڶڕchYhQCJH*aJ3hYhQh@CJaJcHdhdhdhĆh@`5hQ5CJaJhYhQCJH*aJhQ5CJaJhYhQ5CJaJhQCJaJHhtĆhQCJaJHhĆhQCJaJhYhQCJaJ3hYhQhw^CJaJcHdhdhdhĆ%JJ(K*K,KKKKN&N)N@NPNfNpNNNNNnOuOPPP%P*P~~phNNp3hYhQhw^CJaJcHdhdhdhĆhQCJaJHhĆhQCJaJ3hYhQhw^CJaJcHdhdhdhĆHhĆhQCJaJ3hYhQhw^CJaJcHdhdhdhĆHhĆhQCJaJhYhQCJH*aJhYhQCJH*aJhYhQCJaJHhĆhQCJaJ*P+PPPPPPPPPPQRRRRRRT0TXAXCXnXoXۢۢ}oaHhi$Ħh =CJaJHhh$Ħh =CJaJ-hQh =CJaJcHdhdhdhh$ĦhQCJ_HaJnH tH  h&=4hQCJ_HaJnH tH HhĆhQCJaJ3hYhQhw^CJaJcHdhdhdhĆhQCJaJhYhQCJaJ!HhĆhYhQCJaJ oXqXWYhYnYsYYYZZ Z"ZZZZZZZ@[F[H[J[X[\\\\\0]4]]]]]]]]]]]]]-^.^5^;^B_I___``#aɽɽɽխHh#Ħh`CJaJ!Hh#ĦhYh`CJaJHh#Ħh`5CJaJhYhQCJH*aJhYhQCJH*aJhYhQCJaJhQCJaJ-hQh =CJaJcHdhdhdhi$Ħ4\\-^.^aaggojjkqt%wZzzzd]`gdQo#Ħd&`d]`gdQ d]gdQNdC$Eƀ#Ħ]gd`#a&a4aeaiazaaaDbNbbbbbbbbcc8dAdf,f5f=ffffggghhhhhhhNiPiHhĆhQh7CJaJcHcHdhdhĆdhz#Ħ>HhĆhQh7CJaJcHcHdhdhĆdhz#Ħ1hQh7CJaJcHcHdhdhĆdhz#Ħ;hYhQh7CJaJcHcHcHdh ĆdhĆdhz#Ħ>HhĆhQh7CJaJcHcHdhdhĆdhz#Ħ7hYhQh7CJaJcHcHdhdhĆdhz#Ħ:Hh"ĦhQh`CJaJcHdhdhdh#Ħ yyyyyyyyyyzzSzèÌrT6T6T6:Hh ĆhQh7CJaJcHdhdhdhz#Ħ:Hh ĆhQh7CJaJcHdhdhdhz#Ħ3hYhQh7CJaJcHdhdhdhz#Ħ7hYhQh7CJaJcHcHdhdhĆdhz#Ħ5hQh7CJaJcHcHcHdhĆdhĆdhz#Ħ;hYhQh7CJaJcHcHcHdhĆdhĆdhz#Ħ;hYhQh7CJaJcHcHcHdhĆdhĆdhz#Ħ SzTzXzYzZz[znzuzzzzz⫏uYu;7hYhQh7CJaJcHcHdhdh Ćdhz#Ħ:Hh ĆhQh7CJaJcHdhdhdhz#Ħ7hYhQh7CJaJcHcHdhdhĆdhz#Ħ3hYhQh7CJaJcHdhdhdhz#Ħ7hYhQh7CJaJcHcHdhdh Ćdhz#Ħ1hQh7CJaJcHcHdhdh Ćdhz#Ħ:HhĆhQh7CJaJcHdhdhdhz#Ħ:Hh ĆhQh7CJaJcHdhdhdhz#Ħ zzzz{{{{/{<{I{i{j{y{{{{{| }g}{}|}}}}}}}}*~|~~~~ϵϵ|c|c|c|c|c|cϵϵϵϵϵϵϵϵ1hQh7CJaJcHcHdhdh Ćdhz#Ħ7hYhQh7CJaJcHcHdhdh Ćdhz#Ħ9jhQh70J<UaJcHdhdhdhz#Ħ3hYhQh7CJaJcHdhdhdhz#Ħ-hQh7CJaJcHdhdhdhz#Ħ1hQh7CJaJcHcHdhdh Ćdhz#Ħ#Zz|&'2#xVdC$Eƀ8$Ħ]^`gdQd]^`gdQ d]gdQd]`gdQ )+e'2qmƺhH9( h =hQ6CJn$Ħ*h =hQCJn$Ħ*?h =hQh8xu5CJcHdhdhdhĆn$Ħ*?h =hQh8xu6CJcHdhdhdhĆn$Ħ*<h =hQh8xuCJcHdhdhdhĆn$Ħ*h;O$hQCJaJhQCJaJh(lhQ5CJaJhQ5CJaJ-hQh7CJaJcHdhdhdhz#Ħ1hQh7CJaJcHcHdhdhĆdhz#Ħh7{|~̈́΄"#$eI6Hh8$Ħh =h*P'CJaJnH tH n$Ħ*Hh =hQh*P'CJaJcHdhdhdh"ĦnH tH n$Ħ*,h =hQ5CJaJnH tH n$Ħ*/h =hQ6CJ]aJnH tH n$Ħ*)h =hQCJaJnH tH n$Ħ* h =hQ6CJn$Ħ* h =hQ5CJn$Ħ*h =hQCJn$Ħ*#$~jjd]^`gdQdC$]^`gd*P'o@$Ħd&edC$Eƀ8$Ħ]^`gd*P'o@$Ħd&^`$2Obc \꽞~o^o?<h =hQh =CJcHdhdhdhn$Ħn$Ħ* h =hQ6CJn$Ħ*h =hQCJn$Ħ*?h =hQh!6CJcHdhdhdh"Ħn$Ħ*<h =hQh!CJcHdhdhdh"Ħn$Ħ**Hh9$Ħh =h*P'CJn$Ħ*-Hh8$Ħh =h*P'CJn$Ħ*6*Hh8$Ħh =h*P'CJn$Ħ* \xyz%&(+,-01AoP:&:&h =hQhCJn$Ħ**Hh Ćh =hQCJn$Ħ*<h =hQhCJcHdhdhdh Ćn$Ħ* h =hQ5CJn$Ħ* h =hQ6CJn$Ħ*h =hQCJn$Ħ*?h =hQh =5CJcHdhdhdhn$Ħn$Ħ*<h =hQh =CJcHdhdhdhn$Ħn$Ħ*?h =hQh =6CJcHdhdhdhn$Ħn$Ħ*12чr6d]^`gdQdC$]^`gdQVdC$Eƀ Ć]^`gdQABUhƇLJɇЇч8de괔teF&F?h =hQh =6CJcHdhdhdhm$Ħn$Ħ*<h =hQh =CJcHdhdhdhm$Ħn$Ħ*h =hQCJn$Ħ*?h =hQhGt5CJcHdhdhdhĆn$Ħ*?h =hQhGt6CJcHdhdhdhĆn$Ħ*<h =hQhGtCJcHdhdhdhĆn$Ħ*-Hh!Ćh =hQCJn$Ħ*6*Hh!Ćh =hQCJn$Ħ* egrՈ߈dstvi܊sˋt'(+p<h =hQhGtCJcHdhdhdhĆn$Ħ* h =hQ5CJn$Ħ* h =hQ6CJn$Ħ*h =hQCJn$Ħ*<h =hQh =CJcHdhdhdhm$Ħn$Ħ*?h =hQh =5CJcHdhdhdhm$Ħn$Ħ*,568DRpr{|#%&b;<=G͑בoooPo<h =hQh*P'CJcHdhdhdh@$Ħn$Ħ* h =hQ5CJn$Ħ* h =hQ6CJn$Ħ*h =hQCJn$Ħ*?h =hQhGt5CJcHdhdhdhĆn$Ħ*<h =hQhGtCJcHdhdhdhĆn$Ħ*?h =hQhGt6CJcHdhdhdhĆn$Ħ*EȎ|%HFϓғ,-MRSTd]`gdQ d]gdQd]^`gdQבؑڑh!78;E“oP.Bh =hQhGt6CJ]cHdhdhdhĆn$Ħ*<h =hQhGtCJcHdhdhdhĆn$Ħ* h =hQ6CJn$Ħ*?h =hQh =5CJcHdhdhdhm$Ħn$Ħ*?h =hQh =6CJcHdhdhdhm$Ħn$Ħ*<h =hQh =CJcHdhdhdhm$Ħn$Ħ* h =hQ5CJn$Ħ*h =hQCJn$Ħ*“ēΓГѓғٓen+,-.rpaRjhhQCJUaJj@;hYhQCJUaJUj'(h+XhQCJUaJhQCJaJhYhQCJaJhYhQ5CJaJh#hQCJaJjhYhQCJUaJ hQCJ<h =hQhGtCJcHdhdhdhĆn$Ħ*?h =hQhGt5CJcHdhdhdhĆn$Ħ* An example of the stimulus display used in conditions where size contrast impairs size discrimination. The centre circle on the right is 2% larger than that on the left, exactly as in Fig. 1, but here it is surrounded by larger circles, whereas that on the left is surrounded by smaller circles. As the effects of size contrast oppose the real size difference they make it hard to see, and can make the centre circle on the left seem to be larger than that on the right.  Fig. 3. Mean accuracy of size discrimination as a function of the real size difference for the conditions where context impairs discrimination. The performance found in the UK is shown in dotted lines for comparison with that found in Japan, which is shown in solid lines. Standard errors are shown by the error bars around each mean.  EMBED PowerPoint.Slide.8  Fig. 4. The distribution of individual total accuracy scores across all real size differences combined in conditions where the context impairs discrimination. Each triangle is a male in the mathematical sciences. Each circle is a female in the social sciences.     PAGE  PAGE 20 PAGE  PAGE 20 Im unhappy about suggesting genetic differences. Id prefer to cut this part. .IJKLMS9DPQRSTUWXZ[]^`aghiklrsuvwyzƻwwh}f0JmHnHu hQ0JjhQ0JUhQjhQUh.hQCJhQCJaJhYhQ5CJaJhYhQCJaJjhhQCJUaJjzlhhQCJUaJ jwJ hhQCJUaJhhQCJaJ.TVWYZ\]_`ijkwxyh]h&`#$h]h&`#$h.hQCJhQjhQ0JUd]^`gdQ% 0:pQ. A!"#$%'(Dd .(#ik<  C Ab'Vv%i275s'Dnk'Vv%i275PNG  IHDR XrBsRGBPLTE@@@///???OOO___ooo̙ff3̙fffff33f3ff33f30 `RsR{͋մs9{Jj{ZެAj̓bZj{j)jRbZ{RJZ)RՃʹ9 sZޔZ199 )bZŴjbjjRJ A9sJj9ZZsս{bbARbjj99A)Js{RsZbդZZ1JմJZ{ZJRRsޔJ{AJR޴AsެbjJR)Zj 9{RRsŤ޽Js Rs9 jsR)1s{sAj)Jj͋R9JjsRZjRJjb 9{ދ{ŬRޔbZ{b惃RՃb𠠤+ $IDATx^]sHE;,[$6@7IVLVa @K_ P`[np@V@ Q| )Z(ȸ lG1^?)o4w߿O_܁ Q lvqo]z1( { bR` ϻO>,nl|P?(`V>"kwcSdϷnA\_'ȏ__#:XX+ F=s9 U֍A~{:A'bֽ@^_otSDgZ ?I(|62UlO_^bo.(=J ߋu 䓒Ťw)[MPU % ?Q@xWq  ( ) "Ʌj j_Ix5dӯHraWR@$0^MYm+) \6qI.WS@V8J $ƫ) M~%D &N"Ʌj j_Ix5dӯHraWR@$0^MYm+) \6qI.WS@V8J $ƫ) M~%D &N"Ʌj j_Ix5dӯHraWR@$0^MYm+) \6qI.WS@V8J $ƫ) M~%D &N"Ʌj j_Ix5dӯHraWR@$0^MYm+) \6qI.WS@V8J $ƫ) M~%D &N"Ʌj j_Ix5dӯHraWR@$0^MYm+) \6qI.WS@V8J $ƫ) M~%D &N"Ʌj j_Ix5dӯHraWR@$0^MYm+) \6qI.WS@V8J $ƫ) M~%D &N"Ʌj j_Id@~}}IE 02:\Q)d%fH!^(ORbBډF1 1"n{- Ylgıc\2 έGB e{Z2y]K$; oe"2}YVLYY$i+_53J$e5@R2B)I^K?!Ѷ>P{lN+ CsPȲzHF@N&&2l-UB(dU=s R#Bْ:e NCR4, DŽq )`dZ Шq(dZZ π)dZؤv) OY4< b *d6X Nw@&e q^OӐ+Sd6 I6Ad䕽} OCh"wJT߰d6 G_KӐ4|h2}ZGLj(dmOCe M@؆#BM&rR\RB ºg Yo)1l>Ӑz_ H-t=k\i *iHϜ ºg a{, 0z+;ci1<Ӑ:w= ; 8 pp''m"^ M@7$y.?ȋtB|S˭4Vr)oU 9*@: U"ZV$OPDB HV Rh[ߪB@ s~+ [U("ȭ[%LJN ύJouݼ @^#9.)]JJG*9U1^xQ@}Ř]%ǫ" S;&9J|S"TTd!Urlh n[Url^*ZtM&a..D[PyBlƻ FbٲĒTɱ'|bgUr4 s69!dN%*XoX#UR@&\ϐ*9W6 VRlO_邯Ia fT%G{٘1XqYGH{((IJ'֛3p &+dYhm1_"Jf7`hYk^%ǎ; EH~КE u Auà dMVo(()$^UϚ@)h+@6Rev ]{~|o=L9ź:<,q].UnczNY%Ǟ )lX]@f!k\r=&UrN 7YNw;u{r?At'#V/XNNZ%Gb$f_^v+V I*bQ Ws{Db rkxBe;~Y/Um:s`$!9ڕNag.F *.ͣJ8 FwVu JFH!!B"X 9{F} iHDu.Blb)BI,fd _ڕ*MHL ΟcBɸvǴMfǪ\! Y4P%Ǧ@7qiSצ, E#}Vt4lVO u2HBC$G}BR/ ]B FZ9^"#'}&q+w8E^",JM"+9X\n+FϪD\6sЎedB~gUr|>#ъ:Cڰ)fցֻ9#%9EE !'쫐4!8>˳=n&FJIc$߉)|"iM~Ex1SeJ> Q1 N GL)Ud]9ufd׉!)EӚ}251SiBrؘ2wc$vTn&Qebj6nywW~NY:dэ h4P d`&K&c"5<{I%)GS[^a|yʦw LrM@1nq- )19. RQchg!'sނ =@Ⱥge'$;s"Bʙ|A@S~݈S6dDY2B}ȯ˪}~K!#ڷJ9pL#Rg\2lϘd}QMg!1ۤ e iL%8DFoQ E& Ø,<ba3"9pLRG/R\LuPB5 [Ȇ&}O'a!^ qs+7.X@a1Ad # øʍK=WwӇV'Rt^Y- ),a6I&5מv<{eCޠT @B387Mش^ 롱2Sx;ql{5k[k^b<3Js b-k6 FzHbW!LK r%@Y@:O'@ר !Z$uل|(Q3(s ;,{MT8^Z~ ;THU@dK '7$! ww@t9C ݑgU#3ٲy57HͲI&D0h;ȏEʛ%|^vZM/) ]JHo%@S]ǁs;Z ]DKgO=\rqeYT"]&NvѰ["9ꐗݧ'e_df2W|U'O-Kޔ,6Ȟ|"餮[&FNZn wwT*19=j rlyٓ RV@zwX"jɌĬZݨ9g}ݼSBZW-s;vGGvz+Q  6$e[ @>N~r=@̋۫vA8ݼ\ck^k]V^|E *$q VK?-*+?n `YXm 0MVͿ#0UmؚT!nr-@L +/lث[WϠWAz/J lQd+mW |uu wHQcaW2p-꛴}?-=0 >@pI0e0ifp2Л2ӫ26nEv"[ 3lg083*KBR@ 4;%s98 B?؇x#TAV[w+V›m[(}mľ2c6d7{Ǵ?Ԑm@_0ʘX >@b{ q2bK1AN0c)v}47=(s*b d-ۼt$\.sv&7q7PnŢCYNL qyE&vk &J LB.91^c9\4tUs7$]}f &{]]նj*"ws0j0uU@U099] =;$aWTl1 'VӍǠđ s׺a #niM>Exͩ4!˕ U/c?1MV5RfJ. -Jza//nٔ"sOHIOH//ejXM %="^/Ƈ `I3hs+{~TU9:JQM)'7%ԧ2IHJ3ʨR^.{Q/xzXA٩J{#nMh׵&Oew­NaX4YFec6sؘI ˰2 签n_ R- P3 WְFʿh rFй}އm~ld+)2T[𩗦<(üf]m9/KӴ$@2 D->4@D|Şҫ&SFQo5i=R5^JJͨ'Әi.y͝}k/@Y% !RR*M#ʿȢ\@?r䗍"Ό Ʌ ȅG ȅG ȅG ȅG ȅG $roͣ[eP0|s7/ڌWN in4TJR rGBv4 x1q'AB!ڶɛIr9B!^o۞AT!{m1E i={*9:HbJUK.zd_Y$DQ[-vlc $^-wߌ戝 ҲU69pB$D o "H!T!EC$kaDq 8 ")~d6E g h!m܉HB[[9"cv R18ED(d\Erʸe! YvYY Γ>#sadB]/[ ]\?=BU 1YPAZ./EV@J HTAK+ KsQ$ 㥹U H\T*ixi.DĿRz4U@ _Z)=^* Q//EV@J HTAK+ KsQ$ 㥹U H\T*ixi.DĿRz4U@ _Z)=^* Q//EV@J HTAK+ KsQ$ 㥹U H\T*ixi.DĿRz4U@ _Z)=^* Q//EV@J HTAK+ KsQ$ 㥹U H\T*ixi.DĿRz4U@ _Z)=^* Q//EV@J HTAK+ KsQ$ 㥹U < /ʠ l+p?P P`G ^~0 (IENDB`Dd"8*BBF  C "A Ebb2bbcG2,e\Bه[k(nSbcG2,e\BهPNG  IHDR2qsRGBPLTEJIDATx]n9҅]+0[݇zDF%[2IF|A+&Kn϶S7r<&r?vi}o9LpI},~QЛ!=<;]~wGޞq}wt|~_n{7חy)}->'w6{V7ܿϾϿnf߯_Ồ3Ծt{{{n=߯^l,{G﴾{}'=Kn+y'߯Lzwp󓘜( _2/9cyI/_/K%K%| _/_/K%K%| _/_/K%/_/K%K%| _/_/K%K%| _/_/K%K%| _/K%| _/_/K%K%| _/_/K%K%| _/_/K%EL(l"u#n3Oƭ GF澰{]#[ި ӈ芟13Q#)ý1g nVcqHfn{dfslķzt:XھE(*:mKp6JxҾ:qaߊm&q]}Fm\׷wcF|x.>7#<:d㪾ճۼ7Ll㢾8 ؈/' X\p@7n.;fMgx^/S8{6/GFj\cF|bP(s7[#r<//pjF|_|"N8m_|][/////#e^Lz6/////oxq_|_|_|_|\k7]_{@_Ɗknc^8#m{C+V[7j_|wZݙN_%[=nHi_#o.у sϦgA⋯x/fJ6[w3 :oހ:V?9_5F /;/ώ/wF]kޘ5OոyxGBX#TkߚC2_õna|ev7&|=zΫQ2<#!qU mSeÁ/LИ||0z^oXn7Fm\׷jxGBp#ԷbxcWMC4~|޸CSa뽿zxbIF|=or>?po}Y!֡tj79cD4f_r{4j߲A7Vx3nqoY59Ľ1k\qJ joU8 `1\k`Sվ?41k{/K%|Iߙ/_20g| _^6h29h_F|n\6{fmFv|Gۯ9o=eeQ:>ق!pmX|]ק5ʌwo;x3}W"}f9+[󁭾Y -q(\y7ȼ.)uz ^({n[$䮱o[1;'5v-xQ5N-k%>ʷ7F}ͼ#3zm|ϙtmrZ6 }3-|Aˀі=}lV.n۳ܷx6Qjg Ͼ.n5v-FWWk9.E׷Ncߊ /W}ַfsjA|8-Ql;W |}`} k|,xw>%//ː8_-/n7}[0)|_|ߩ|_|_?_(_~a@_|`ż~_|_|} ޟك_|+f"}| }k][_mh_͜>)HϾg|WfF->2wϟ^6)wFBxt73}6*]}FÓ~)S{#'~@|&ir{ѷd]W k5l}+^p0,5⛆x%/(XwE|e{|P.VC1u16 .]x&>_m<vy/_/K%K%| _k{]7Ԙ5w.zn}빚>G @ t;β ]o 6kH_5r_Z/{j_ תﻄ_e }P櫆]w8: \ Ɯ-qڦ_|ek!h\quljjx^-AU JhU-S7+%_|O[V_k\طbv o4FmK˵?Qb]({zߺy'7nF|=oě7WAoQJ{}uې{Bzk34J`㢾+?cF|H {rmo4ھx_|5n0[1%ސk]x??:LkL}%qQ_U}1qF4:qԘ%_|;NM7[Վn\79c }6;{7.[8 a{x&g΍]W 36J}s/ԑMYߓGkggiFfoRZ4{ ,#4j!7v4m@|]XxĽ1nM7Ac_/K%|%| _/_/K%|%| _/_/K%|_/K%|%| _/_/K%|%| _/_/K%|%| _/_|%| _/_2/9kŗ79o6Bȼ?#;t0{UIENDB`:1Dd _5$__0  # A"0`%,0;@=0`%,(rjbZx xENսa_ "!"(f*(A6ٔ2, 0 >u (,wNwuܨ8}oW_թoӧNu7q~jt{UqIq*ڧ;p; : :΃XYG8GV[@:pb q"E8Wc&(e?ݡSGݷZrh^唂\~9!\YͫĘ699򺜝\E12`j9}}o߾@AZuQ#.L1s Au|8:`]3_—)KYXjZ:IRgMLt.(_M\'AULqbI2ՑaL|[ENNaXo:IKD_uc ;z|wL5Ot@Ǟu/!~Wt4Z?u/}BɾL?E8β~z7tleRϔ~Fӓنhn]2ʔ'Jx8}ASr^tXTtEPHܠ߇%N?PQw#tDcG<Ƿ#CXG ݬnVnN2oocڷZO2oOړsg_Νte~oǝmF奔_^k_'_܆ Nxdh;c19ǘ?x{7ǘ?x{վV˯!YN2o_$2d޾^io7ؐ:ݍ?ce &r?x\}BJV\! wArlKev%M7cױMQTEAnyk:?VUKo{FzYKuIYCq"܀猕~o JɆbQ j}0V w3Vsw|3jVsSfLV:y:::ɘoҬb,^udX]uϡC,3sX*3r}0(a(OGyhV1[01SP3AyK:yLgZH:}.{YOBG3c(Nyq~'q m/bo+僌1"iP>"bafX~JaMjfPބ 025gۇjX^)z`-*5P֛}BQ zcyc}aY=`úb o: Sf tQp â3c܁ҵnѬq&ATloXߦ5 0LfAc `>4a=K6c%1TWBx /B]ZN/c:KbxPn]Yp EqR6c-,gX߉yz 8N lsoнwd',rHa>3-Ez~rFNAE>^#`ou_dw[' XP>ʭC CsU}1X3*(dF kiXO܌fRlg1If&m&ge2z7<ۄr=Zr-́kR/ b% ^ kJ2(oƼfeYee"?c_hsbCDgqQ̰hQh^3?Eu.1gsP^2~}~b 3ַ(xs<|/ºq~ul{e,|.a'4/CuIw9Bq/cŢc@y;wjV SFmǺwRjٶ4?FƪZ7cI NZu띚jZl\X( #G ۇ$ќպoZܤa+ٖ`(XQ^B2/kV{U-f[4B܋-YXPdz|T3X4Ͷƚ,ƚ֬3²Ǻٖ`e(?X/< 5ܺj5 9ˑo*6Na7QX[QKs.Tnxguy6g}c(QӬuY1-yN0c}04Su#j=Gc, a `^<1] 0)x`af[3*|V%ZOJhV@_f̶ԧ/c5D7c5@9fܺN@}̶'n q٥aOKbi6.ejݟ%1ySfjZۯ}joobLbr5n췧[WIꯒ oZ;E,w ~o%DJKkvWZ[ۘxn*'Z"MmLӧ룋:*Pj"*X仰X(%j'*΢*ʫ(,հ/B/ݫ.<-cQ3ƚct]uGp_.pa->kd,~kYGr@&v˕cZ}v1_b:횵Yl&Clvâڣc>٫P/r^*_ S*cQlEƢy"^Ҭ/[HKBb1yY7Gx\:'b0[^s nZOS9 ̧jVa$K8yL`aQeuF sTal*ǯWnò: ` >PAסܗ-׬zG L _=9xk>bYv e+U@wJ>ⱹٱ1bҋS6Ь;z'ԗwC>?Ub=MuU@(/@E9B+0P1/)Dq5,ʱPP> 1O<^y&c4ױpi galbw3x75˟aZfͅa|  ՟327kʭ aI' ;l鸹":~ܺpB6+ǷɺfՒNjKgm /m/=I[ Z4d;ߠhomrn.4AcƧcIz>Ьhˊhq+C[Yðnc݁r,c.ޚn).K@Y2dQS2,Z.Xg^wہ"/ k0t?;~#JS:7=1'G)c0zgg~Dʃb+BIyİfIG[Y=׉[xN^ʻbeXS}GXk]}MRnɵ ^ k^s?RJ,RC(Dsz?zϢ8,鹝rO,O>-W%2S_{N驚/9IjX_KqӇ|q/h?_uv=71T|}5K{EA3"Valdʋ(wUZ݆N8"ʩEsΌgc]C9ʋvM8[Z⬭as:>nFgҜ1o&iSՍ"U%h}>|]V2V'ArEwuj(nU $2V_=Y Tn]=1PŋA8+L _g|=WIܺqUSLRudu}y5iPyvU TPsT-â>ոMPzE?N3[wxCUTe fEy3bgA45}ETiL>e0ʥQJ3[WR|T)qfN% ](_LUX K8L`ca*x|V={y!*JE &X5޺P]T^\:5iRa>_nE;|^ǣ(SK!1VLb%ʟe1D+FhU#L14#c 4F};c9+Fx.>yp#`7D]wW(co1Qߣ"T=P5#^j^nV:{I`56 qsuޓld뼿dǬujOO6?euZ:-`y_WK]oݖvGK,SNY;E2,e\q7q3KSW`t{uӬfKyc=ycY1罷 =  w}rΰ /_(SS6KۢUKZ{5Ew4r{O4һt4"KgC> ןj5/q^VeT_n;t|QX :N۹VA=xJbr~<>}:5&븊r}R4DELTΏ篟G;ooqZP縨_ߏv|_gZO(wۮN 'sHw>ᰭBϭ?ᓱ<X*x~!?v,5,IͯXsm^)}EsyO==$SndUMtEr~zמg=Y/}doȒϟ5eG<f9s^5z'Q.ZEMLTΏ#-E&*yGҵ?yGNjp>=N=#d̟гTLy,~"/=I.?!a} ݗH5}})DӶ'5ڶۨMlQŤgI}̃2:04sDmy`6ä1Y>305Ofz72f7z9" zհ11Uz*w@`eb-ǔ}X7ƛ[p[Ŗw>\JC>N=h+~ SW\¬O[q%}i|•/RH3"4i!"ղHѶ"؊V؊TVZ"Eۊ4c+RNPJC[b+R-[jيm+ҌHA;A) mEZHlC89;JŃ%V{lodG/a׻lƦm^vzl!{7ݶ V20؅t m茞^ Nm(|06 ~gi~`g1e;@3zm@.ϮfFfC> >8l_[N\HJrw*/s[to {cp><^k빽#ɮv0v:mbo}ys߶/ȕe_>?*};>7[88683]R;_צ8N ,粼SN㺄Rr~Y%L UC}}rjsybZx xUյ׺I1Xۇ  bR*P@B qFY-bˌ CQ D! š(@o{&sk=w5 ?Z]ꅊH1(?J\%-H.5EDu[_7ˊ"îW~}bJ RjҍSBN Y}q0f,.EJi9OQ9tu6Z!R_Za֓ n=(6nƚ(dق[BUzAdMT=ranK&a- kQD&hUܛh*DEX)Z4gEVae֢?X:+#Yj`-VZ(=&6XVPAuux~| 8TRYEAk2.J"Umx~acEvaC+-#}h^wrnز9 }i3 [6C6lAZSf'Eذ) Yis6lnBl{ ?6SaRg3f}l؈٬yVJoXyh"#Bf5Iy؛оFI6NxDh_meS"#Bim^6gb'<"6gz<о{\_o\A'd+L2M L՟A7VVOf2[d}̲Seӡ*st̵85edV) LbUY"bh+Y& Jr+\D^]%?0DADXxGQ>ߗ? Vlw)f'R(du}v˳GWdWsVKvK~_ZY X$/[64<6y {neyݽ&;| ~ge?]xe߳ռ?ćVpO?Џql wxww);zwʗs_s>*n\N❝͇~wfjHf3Yj5+NjAOgW~MֺxiЧ [Yx|}=F68=<IvVm(A0 "ΖVNϔda#Ȑ#m?jF7BkCQP'XY@/`MO^ruN J5 ]0 B= **e:UiTõ2T%smn_<6 .(,YBp bkrTFˡW+^(TUC*_eeMd=m6TeC٤d& OXٲ~g}B6Vvlyq;T?ce<=NwPy}d=K]/RkCUЯP RSZȎP Lod?{ߏ9FCOCmj}oB v5p5ߡSĮwS_b/[͵;xջxwpvB Q?>Ý{h>Q+$mn,E-?][ ]m+;wqf#[K7w|=#{/(_ >N 4xJjrGWQF0\>R<㭋|{"x#C?) d̅׋llN,pi&lN7 ;Nr*ýQ'N5F_&rC4]"z+`"8Dl4ae n# (t;}c#XDXt:Dr"/:W[Jt&M mxsmc1 7YYܝ3!,7C'0w p0#|N~@A:|~| = I5 z8~卼?=@'ρη@hÛӅ/&C,^l >nˠWX፼7}A_]cf *y/z޺b/FoKog%׋Y>u?{'"Oиױվ>>]>:ܣOиA>䞛o'?O?t,'ښܓ䦰䱹{;꘲챹31"1֢LsʦL{OqC?̳ۇ~ğ~ݰ⧎"g$<{FL&H~{^[ޗռ+u?7/6b{0gڟ})>h_Vj9:̲h1?FS*;r>Gu;7* ~ҝ^7} N+MEZZUznqUOz}qnTqxU~qe٥VseeʲKc n]~`}]}"f:glVVvLod 8 K]bϘdLM\se%Sr K]bdXdT$}iϕ+.%.' %Q;.AvEb1ϓtyy\< jV֎ \g#|;קmJpߕ{- 0w\)ٷmfh`y3\)=mOHnX́p}ڦt1z&S8V96'Ϡ ʬ>cM7jO۔fn6' %Q;.AvELvibקmJKig]d "W"ȋ.>mS݋6' %Q;.AvELv$i2=xd "j%.߷AaJ ,?[53;wȘCnM1*7}TWNcU+ tSֲHPNcS|[̷;q^b//jUg3GW++ZkYnZH_Ҭf7Y zµ8O,yCjWHpuM3RlH |[a YOO&Vs\V%+GdXc& @C1 C5m\cG ˇ,k숷|ٷֲϟ d]X:xC{Ε{4U& *j_{ϋwjϦ,U#cgYTݭimgsWT!R.d+Y qa1)S n ]hn; T[B;jX>,g+ݳҼ壽<>rf2}oh]ȧfO>]}Ob?SXڧ_jUӺ3X[4~=w7V5HFU _~ T/,ǿw' dWkT/S&02NjԲO4b\7U38{ۿ{}M|M|;}|GZ;ѯ9 EPAwDD%SvIKfu9F};'$GmkPՉ8TEMDd 0  # A"FM.pcۣ"Ml@=M.pcۣdEO4p&x |TՙL I a$" D$ `|[4Tw( Jb P4j"]%lխ/Hֈt-h+R#;TWE{ι7LIso}{;9sqcamXec cM7 q]Xd F뿛{ư^6jt\1|ugP">6hcc0zrzY|\P䖺zHzʐ$ףȰ0>,qGxv~ #PGtOY"ڌa}꣨~ͷ?fT;m$51& q[>|Iz͏Ջc>FճoQGS%s_|c65M }mCu!ql+7}w-s}sw~ a(lC'8H.25&ϠlUZ>CϏ/ȕ`F)5:{ ~ q\ͅMԏ'Qbg2y _u=gɸ;RGߑ6~#o?>>L>p? }۽;k`7a`ykyafF1ީDͦ6=bfbln}o3kdJ%OEw]TaŏL@6{h6Cجª`7ֱ϶mQw#񱷲Sٛʹ7w,ζi7֨%˱_Ϭ?ͳًMCܫwش-ʲ)ȭ\[&1Bm63fBҎuTSQYxϴx1^}VcvN-S 4{raM Al߄8;f[lfŀodqŗJ3=w֨dJ2ŏ`P6j6﯅Uy08,&1ׯ=6ׯǰ-%DzSG<}_@fYfUx+E59FX3 #T>{lƀAil+d[k}ſm XDdTC;~^]6_c?qwrR鵶6Dl $m6mwˢxTTSOG}fF6XlvhBZmzyV-ʳ dÚ`A qwrR6Ell*pE֠dO`Qll2ŏ[ʒ~E,6:M_{6p6G[lr(4%[a{U2H(=xߴfA <+b?IelQ5(?Xx(ѿx*?3g,7f߼%} {?i^o5šQnfx'\7ϰ96XPV:Wu>>ϼ_]Xfu_:#^:c>p6Zlqhy +o~㭅Uy*d\_g  McfΛ6G vd,3T:-L8,J3 B!w?r|u?H=^k,Y vvkm|>BrAH)}m)\;ؚ͗+Z)Gr_~x7„ .!?{:IuoWkު}m}L ׵:u~1W8t^h4tr^Gk-#2Woa{>pQ.8?`8.:;nDX;ҟr] {$y_^tH)-mXR6lDKƼm_Wcn!}+Dޖ6;4|Q'LU}&2u o9mg';LT#E> g};}SKm]RCڵ=OZڵ^ڵ1׮uEƹFn][mؤ!:w26/wUe,ttCIg?]r> pĩm[g\r>Eⷡŷɲ>8u=Qtɏ/Ծ߱>µO?ܩ㴥杶?װ:>?N8'8ٰ·'OƳﯫ;< cNk?}9 '$?k?\'sqE_':\ىӔ?uUNl:3n¦۬[Fg۝.y1t| Eڝ0N>74ϹQf?85ϵÝ<9!Mcd,}gkN:(5b5|)]0) YdRV_X)=a}W+Ú!ܞ#0# `yp)i +լ4b%&+"mo P+v+jV޳Xy'X,nXyL[[d%52+&Ŋ{bX>XyP~X_yLb6Y.+Vx~]~+u7&+gd`ye+Ob%&+}#R٣I U}Z'ZW^䕽XM=ccodym +דuR^yd9+{/g+n3g|Yd%=b^|+yEVꤼRJyVX=c%&+"¼BFWZXުJ+Jy֌^ΎV66Vw6YɌ|=$pjĹk::)ӿ 89׬,JMVDde'Vk,: 4+Z|گԿ,^XYŊ&+ّY!#deӊb%Nbefeʢ?? +4+b%&+#kb@+ t+Ś*JO5⅕b,[dePDVYdX&XYb妌RXceMVD<"+<"#<ҬYIAcquqM3e,բ椾˽3lךLXa^YUXWdlK^a~y[b%~_a^6hkR)rK,XP\'Bx+91ܠQ$'LI8S)vm,(Y{i;5\ o6ίk}y)*F*3nVeN[tv>A.:v>6\dko۱O7I]?Ka3?X?YƩ]yĄe-5%-sW#])tk|]9r+΁'3(6b1"w{:;kA'Q9|+L \4g>JDx*?3_ :\Rݹ4 Nyy`Wn1,?p<(?0bPq`rX :lqwjJhl>131"Q/n+g{ϲꕒ=,O<}fsfbX~ ?;?098pev9Vr_8;5l3l&Edp«,z&Z8uk{Cs֨dJ2ŏٌ>WVwyfsŰ2?ÞGy08,CAffs-6cfl^`G%}VԨzNtW~#\Q"TSQY8n6olbsAf1,? ÙXu^ T6/l^e͔>)6ӔVnPlBMI?ܠ4͇+6&TؤLٌ>Xwyf6-Ȝ Ùsa+`Qވ_8;ͫ5fw .9wv>w)]{wE֠dOcn9~ѭk2ϹtlF{;f6/,䯄mBpX5l}Ոcu>i1`o6ͦ fѵMu9~zl>8SG㬻Mo(Ჹ;,Թ௴ĶIJѴǶ6٣ő/?,ܶw賾+֨J2ŏٌ>Zɳ9Yy<q` VJ096Ex:fOf-6}1`3="Nd٢U.h+?泌WJd~*_OoͦgjGrn65<Ű&KQv{a5:밽1l~c(6Sm6EfjTؼbbsOWX욬 6QYxތ>w\wyBL`2[(AXr_8;8l^e1`33q&T؄ Jy.l\؄ Pa~2g3\u}~Bwd$l:l!ʫŚAXr_8;Mo=6sbm:LS6\dN6&dsRi&pam:ل I]xތ>_w|wM3 68 [c(ͅmDy#oΛb37l|ruM)khF? uXu^ d6lcwԹu(Y|c:Pу/e)?YW*֡d~9x*?3?>aɳyӱ-6"<*7*P:kxQh:,]( }TRQuRu[͓~aҨÙ'FZ:j^\@V5+#8!YRdؑb%~:|U3;)V̷tԲfk۬t\eNV' #e39pfwB +G,0$輒Wxy˘Wd 4! fb^ɳ*ωnQ1'׶Y颹W a^ia%~:ܧYa^ɳ*ϗd`+ ߃:,Hrtb忦?s~Y aeL*OT+{89uVfCAeo6hji6#۠~V=m5s "+l۠Vg}~Vd`ye+G ;t\ҷշPmy=9:d^yM;Qy%-;Ӗ3-^~W5sJHHb%~:ܧYa^1Y8:-Kpdt|f+]3סm d}[cJN+c ҂ढ >XanJu(Q 9~^_aA#ץ..5^Xfe-VfCdeeCZS/~:$+4+O[lubdnQ)͚'uHV8ס\ȭ]4!YYYg5Ӓtl[g)ʵ넕 [%@+>t+4+ ,V{yVQg% +kV~`&+T"T~aX>9b≑OɊ}>ΉQ#tN>!3ѿOC")<;jxEO4D&x} tUչ-PANIx(hxh}^|[i* dyJB Iϣ{=n<S-yANAsZx}ߜk&k:a|skkJY^p}''{qtW*lH 'G'Sͬv?za`NgI ӑ-|Up^s㕕F<km[g|yS9s':ߞ pӟp./[ptw#+VϹXwỊ{w-t+ CczM9yr 888w[N mRs3a̡k'Nx'PX*1~7:S=m/b; ^yӭk}q3zi'ks% ofsnjFԛ(pnyYRK/վTRK/վTRK/nO)8 2W]:MDW^U12Uk^]]xRB_i'vWM'P) I#_mWOT} aZq"^qs;)mEx>9)U^'CQ.@V2N%l1B(1ʕ,a! ۵e<ةF` u)wS_ع/cбKBT"Xn.DBNC7~"Rʼn#?^.H6-%tŰ"B]tZ$`IJPΆg+v|Y(vJQ X,ڕ+?.&ԮeⶾrxS'jLJ?sp̆Y#?^vi|..Q^ -A F, lq:^%_ʃ ַ ׭|}ӌ-zwmcK؛k5^qsӯ@tA`:W:Y{isKC cFct{iqalcc6_)^>ɏV7^ǿ3{|tgMǿY_?W]iwyv3 a+aaV;!c1/MZxlK1-Ƕ4X=1wSǼݕMpys76΃ysA׼#?^+=zi|..Q^ -p\._b4ql4mctXt;o2:o^k|>³ܶĴʳ尭:ϴѼNp~&TH}ԋq"^qsӯ9 >gSx>w US(Vr;i8lӇ8s}L).Y?}=y>D? o>-댉@zu^^l8_oW?A3O3xagi3E·) jwm:ʧitjg],vWwv}[yBuB-Fqh@y¸[P#?^>vWiW>*pFW+6S*O:3*lʧjX jW5AE$o}{wmO=3mfӞK}h3O]:^Tb҉H݊|Up;5K"u"R7ƉOd{őuwrFM1 >Mx&hhk@E"Ubk@d0O)ږˀF?o&&~:@o$oƨK"퐉?,=ԇH'>G>7u~.4:vaU ŊP.Vð FW>2~[>(/u:OY.AY01Us@UVebsp`9tOj$ru[1-|}&~͉DΑDΙY# Ө@OTW='a.ۃ9->ݷa0_m?w| (VrΣo̗TJѾ X4(_cCm9;@ô(9F` jXCm9wC4Έw -tSĿkq8l8EyXX؉1[ycM;wkL.ob0۷8ܭW.*E2`ЌÝwƷq8 'װ`( hhXH". Q_pa6DҲ vX KEmWl FߴnnTR/ۤDmafht.%ǜ+;QL+50w2Nd}"+|nx < 34a`8 \ +V*,v0 :UخorU} Xk5Bv"6K;@!Iaz3"O" ^?6mԸ"r9'>>7ɴ< k0cL܀m&|/Bو&kvv0ۈ_0^m㏹|x|C;@aɿ?| olo\ku&т﹁q}G>7v<톺ڭF2p:jV-o-bj[+Fk:hjmRN |>j7!v`J/ά6uiFu1G>7v7x w{_/_} :d[-ɷ1rx7ڭQ%hՎ|x|nDhwcq'J QZ>5''>G>7v7zڍhӮ|U4r%l1j[_ֱ*lv|kP^K8dwTndhwS'7Ԙ%@=֬=I[b֞Rg,5kIZ[ĵG>7v7yڍ4BccXVqĎZ)Vr/-cma:Se2{_g}uMcoTws'WKZzp Qz>PKZ֟֟fOQmU0 Q bhV0 gBajk~U_ئ7%q25 [=n@_'"qP:!?^[pk(/Tb/18 1 UJѾ X4c 6svck8(h'g  o~H". Q_p᭮y>gKbgl.c9s;P{sY|P}@9VYҪLҮN(pWYϺ,L'xN(of`kw LO' ۰73K O%zoW0t?*CW& .yӐ9=adIW_N3wsw>"|;m3?/;]V2w .tJfoۯ>Fl0@ eɢ(G1o拠(rE;r1G/molB&絈i\}rz_#(|fXr|Yc(~_(rw?no'w$Tsw.g RAA2(0]Q1"W>NI1G4 ‰8_' <  ȢC~]!|ƢG;Ô Ρzr89߇4q|r|ˢCy~swuY"avr%q#qYqH4t>O#OYpp }"-G!fQ#îI9Xrh89Na?cOOS}{_E9 y~5rGI͇?/w\V׋cV wWA2rwy;"9bN͆?/ i~ư1`3k'-2a 5ɢրr} F!˭Eec4~!]ravr+P+ a;@_ً>p?z?Tno,:'A(E^N$ܱqkWϼ܏~ꀬrr|dP -;|-k_#lS?+]#s8̊zg76臻ЯvL?ܩr|;eQԡ[3(rE~#o N--#(v'v|oȢ#lk?\|;fr~~HwlGw K:8Q{]<۬(rEq5"EG4)&z+Mo8d翍*7ȷQE9 y~0rGRa0yq$K5b[GQn5-,ʷQv cVokѶE|<_ZB='[] !B= <}> q,,o}6uN%;g@_\Zqe+`Lbz[/،c߬翻̴4ƭr!/ʳݪrڥ(в2Ƣ~8ސóCø򴡼¶-bpݤrڷ(вG/'7&?&噳2ø'<~cCp8-Xno, y>3(󑿑䰨Vmp~ a?}8 M'}REQc4?\|ot;%5暸&<={pNWAeQԡrBCCzƭ!$(| kh9dUcCkE|<'gkxCG^ qX53ǣC/30GrdQԡsG&Ll׸u~@*?d!336ponI>[#p_;-ѿ"9[{M 3O?OL[D'OEں@I t>~K` 0:/giΧπ1 XlY{1|(Dv_>$皞@3ƉeE +_H/HA' ]#W&7а FaET kZlj\ h5YlSCDnj7iDfN8&k~/MԈH'>>7E<)  >|4|wT,wjHAhu`w;Vߪm|6SCDށ]_!g:~U 1N/^q*OG :۠N |n;a5|ST i57ms0hznS mOIr^c46!P2N _4hB_!P!濈$L9 N γLO&*vYW` ['is pm}ǧ%U_'? W_+ H纤j"Oq"^qg/iI9 8By$88xg%vYW&5 l-NkGfhh!ێ9f7 {b:~o͹ȹD>+ʊ9{Կ鰪W_8g1܀ṭ?堳M?7(0ҷ1yٽڳݳ:+"ޜ_"贡>8Ͽ,66E<1ikuuootKZ+ЍZj[+G֪ `Fy3/G{bRԇH{xS&>D8l8/ԯ縄듋 Xy|#X "a BT=}c9jH9*rL?٭F"͐9ئO {r8U`4J)Fç <]M\!!+|nzrRt!jRۅVB._XNiNP-9k!4,^ 6C>`~j,x( 2Oi@i85| 7!?!P!O09Ct(<<؞yC+DqZ\dG/yf\ q>ZB{>@mlQdq2 3΀wuČÜi›g>eP qǸ!^q~ok_a]s1fb,bLgg`΄Rr>_Ovx )t X̌CLmg5|}ǡOenjS~ijj$ / ^qs_u f1yLOgӁӡt߳-]Zs4Ju`ӳ?4=֗ ~=B MyE'Myqk_& CQΆe26iE,\Νv?sbfaVlo(^+>PfϾRG΀od>[6Z12>;r1P#` f1q)>81(T{S?ckSmz cQ1Ls 4 ]somفl̫5 }{5qjP@jƸQ#?\&,.ǻAjx/ARQ$hA!@l ޭ1}q#K0}>6T1}밿[t}׿G|{dC y~rKmCf"_|3jC;&\O+0j2g ,TW'G|` ɿ_{ &MNr8LpOAM9p086>6 >sCW':p8/ r~)`?!!?a?kSSPVR>,˳?,9$,ܳ!ϯ6Ÿ!9?$쇳&,*=:*ϗT6y>߱E`aøŁ!y!۴-,Ȱ*+bg+ϳ·𫱹/ٖ㗿R񰭳l]\ ןt !Oc?JvA.Y$%Odͽ*˗B|m㟉\y@x/碏Or_!`>TH=| =ocċn Q.")c?:.u;=\3dL!GSS1(3ǎU,u!o>`ېƑs9g08$ܱ߳$s{dﻲԡ紓ĿCf*CJDA@D Default Paragraph FontVi@V  Table Normal :V 44 la (k@(No List 4 @4 Footer  9r .)@. Page Number4@4 Header  9r H@"H  Balloon TextCJOJQJ^JaJ>'@1> ?_lComment ReferenceCJ<@B< ?_l Comment TextCJaJBj@ABB ?_lComment SubjectCJaJ6U@a6 (l Hyperlink >*B*phPsychology DepartmentlMDĆQTI|EOP-.02A p q ,=.#F#b#z#|#'''c,d,c1d1r1 3 3559]<u<<<<2?J?_?w?x?B4F=N>NSNOOgShSYY\\]OcIfhk.nqqrrr,s7t2uvwwxxxcyzz{{||%}}0~~@Z1ڂy!؄!ad)*, _`O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O900O90000000@0@0@0@0@0@0@0@0@0Oy00@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@000000000000000I|E-OP-.02A p q ,=.#F#b#z#|#'''c,d,c1d1r1 3 3559]<u<<<<2?J?_?w?x?B4F=N>NSNOOgShSYY\\]OcIfhk.nqqrrr,s7t2uvvvwwxxxcyzz{{||%}}0~~@Z1ڂy!؄!ad)*, _`ËċŋƋ0800000000000000000000000000000000000000000000000000000000000000000000000000000000000000@0000@000000000000000000@0000000000000000@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0 0z @0 0œ0000SNOgShSYY\ƋMy00My00My00My00My00My00My000M90M90H  %0>>>AF N )p!+#%6'|+,,3>J*PoX#anjt\xySzz$\Aeב“.KNPQRTUVWXYZ[\]^`abcefghijkmnpqstuwx} p/\Zz#TLOS_dlorv~M`|~: !%,07:A! !T! !T  O2$ 4Ü_-S 2$Ko& i$Nh2$MR$$~N$=x2$F"-]+9 ߺ# @ 0(  B S  ? '!'t="'L=#'$' %'lPB&'''4xB('D)'ĮE*'L+'*9,'|(9-'IE.'F/'|=0'1'G2'G3'F4'G5' G6'%7'+900??wJwJNNNNvvww     44DD,|J|JNNNNvvww 8 *urn:schemas-microsoft-com:office:smarttagsdate9*urn:schemas-microsoft-com:office:smarttagsStateB*urn:schemas-microsoft-com:office:smarttagscountry-region>*urn:schemas-microsoft-com:office:smarttags PostalCodeh*urn:schemas-microsoft-com:office:smarttagsCity0http://www.5iamas-microsoft-com:office:smarttags=*urn:schemas-microsoft-com:office:smarttags PlaceName=*urn:schemas-microsoft-com:office:smarttags PlaceTypeV*urn:schemas-microsoft-com:office:smarttagsplacehttp://www.5iantlavalamp.com/ 1120077DayMonthYear EFGDI:?O"O4O;OOO RRKTRTkYrYUu[uz z%},}!(ŋƋ|}E-N> R ,,55HHHNRNWWtuww)x0xzzŋƋ333333333333DIŋƋPsychology Department"X]S Z Gf tLo hh^h`o(.hh^h`o(.hh^h`o(.hh^h`o(."X]SGfZtLo  Qaa*P'}f`zLU< =rw@II<II()*?@GHLM3`@`0`2`h@`J`@`T`@`Z`@`@Unknownwap1phillipsPsychology DepartmentMartin DohertyG: Times New Roman5Symbol3& : Arial5& zaTahoma"1h#Ħu$Ħ$ĦbvGvG4d@@ 2qHP@?L Test inputPsychology DepartmentPsychology Department    Oh+'0 $ @ L X dpx Test inputestPsychology Department1syc Normal.dotDPsychology Department17ycMicrosoft Word 10.0@ @h@l b@>Vtv՜.+,0 hp  Stirling UniversityG@A  Test input Title  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~     Root Entry F7htData D1Table;WordDocument'SummaryInformation(DocumentSummaryInformation8CompObjj  FMicrosoft Word Document MSWordDocWord.Document.89q