visual illusions autism

8/12/2019 Visual Illusions Autism

1/6

PergamonS0021-9630 96)00052-2

J. Child Psychol. Psychiat. Vol. 37, No. 7, pp. 873-877, 199 19% Association for Child Psychology and PsydiiatrPublished by Elsevier Science LPrinted in Great Britain. All rights reserve0021-96 30/% 15.00 + 0 .0

not Succumb to Visual Illusions. A Research NoteFrancesca G.E. Happe

MRC Cognitive Development Unit, London, U.K.While anecdotal reports of abnormal perceptual experiences in autism abound, there havebeen to date no experimental studies showing fundamental perceptual peculiarities. Thepresent paper reports results from a first study of low-level visual integration in autism.Twenty-five subjects with autism, 21 normal 7- and 8-year-olds, and 26 children withlearning difficulties were asked to make simple judgements about six well-known visualillusions. Two conditions were used, in an attempt to explore group differences; standardtwo-dimensional black and white line drawings, and the same figures augmented withraised coloured lines. The subjects with autism were less likely to succumb to the two-dimensional illusions than were the other groups, and were less aided by the three-dimensional 'disembedded' condition. These striking results are discussed with referenceto the 'central coherence' account of autism.

Copyright 1996 Association for Child Psychology and Psychiatry.Keywords: Autism, central coherence, visual illusions, perceptionAbbreviations: BP VS , British Picture Vocabulary Scale; DSM , Diagnostic and StatisticalManual of the American Psychiatric Association; MLD, moderate learning difficulties;VMA, verbal mental age; WISC-R, Wechsler Intelligence Scale for ChildrenRevised

IntroductionPerceptual abnormalities are a common feature of theutobiographical accounts of individuals with autismGrandin, 1984; Williams, 1992; White & White, 1987).ypo- and hyper-sensitivity to sound, touch or vision aredescribed, as well as fragmented perception and intensexperience of normally unnoticed aspects of thenvironment. To date, however, these possible peculia-rities have not been systematically explored. Frith andBaron-Cohen (1987), reviewing research in this area,

concluded that there was no empirical evidence againstthe hypothesis that low-level perceptual processes areintact in autism.Current psychological theories of autism (reviewed inHappe, 1994a, b) focus mainly on the social impair-ments, and have relatively little to say about the non-social deficits and assets seen in autism. Non-socialfeatures which are clinically prominent include therestricted repertoire of interests (necessary for diagnosisin the third revised version of the Diagnostic andStatistical Manual, DSM-III-R, of the APA, 1987), theobsessive desire for sameness (one of two cardinal

Requests for reprints to: Francesca G. E. Happe, MRC

features for Kanner & Eisenberg, 1956), the islets oability (an essential criterion in Kanner, 1943), the idiosavant abilities (striking in one in 10 individuals withautism; Rimland & Hill, 1984), the excellent rotmemory (emphasized by Kanner, 1943) and thpreoccupation with parts of objects (a diagnostic featurin DSM-IV, APA, 1994).The 'central coherence' theory presented by Frith(1989) and further developed by Frith and Happe (1994and Happe (1994a, c) is one exception to this generatrend. These authors suggest that the non-social featureof autism, both assets and deficits, may follow from tendency to process local vs global information, and failure to process incoming stimuli in context. Onquestion concerns the level at which coherence ihypothesized to be lacking. Findings that subjects withautism are better than controls at spotting embeddedfigures (Shah & Frith, 1983), and that they are betteable to reproduce unsegmented block designs (Happe1994c; Shah & Frith, 1993) suggest problems incoherence at a relatively early perceptual or attentionastage. By contrast, demonstrations of failure to procesinformation in context during sentence reading (Frith &Snowling, 1983; Happe, in press) point to deficits icoherence at higher levels of meaning extraction.The present study represents a first attempt t


2/6

874 F. G. E. HAPPETitchener circles

o oo oo oControl figure

Are these two circles different sizes or the same size?

Muller-Lyer figures Control figure

Are these two lines the same length or different lengths?

Ponzo illusionPoggendorf illusion Hering iUusion

Kanisza triangle

Figure 1. Two -dime nsiona l stimuli for visual illusions study and exam ples of control stimuli.

figures devoid of 'higher level' meaning, which none-theless appear to require the integration of perceptualfeatures. Gregory (1967) has pointed out that in manyillusions 'induced' and 'inducing' parts can be easilydistinguished. Taking this suggestion of a figure-groundseparation, illusions appear to present examples of theeffects of context on perception at very low levels. Forexample, the central circles in the Titchener illusion(Fig. 1) are easily perceived as the same size when seenin isolation (as in the control figure. Fig. 1), but whenplaced in the inducing context (of surrounding small orlarge circles) are judged to be of different sizes. In thisillusion the central circles presumably become part of awhole figure gestalt which changes subjects' perceptionof these parts.If subjects with autism are able to process the induced

might be predicted that they will make more accurjudgements, and fail to succumb to the common illuseffects. If their better performance on the illusions isto this weak coherence, then a manipulation whartificially segments out the parts to be judged (throcolour and depth) should help the controls to performaccurately as the subjects with autism, who shothemselves be little aided by this pre-segmentation.

MethodSubjects

Twenty-five children with autism took part in the study.were attending special schools for autism, and all had receia diagnosis of autism by experienced clinicians using guidelines of standard criteria such as DSM-III, DSM-I


3/6

WE K CENTR L COHERENCE T LOW LEVELS 87Table 1.Subject Characteristics

GroupMeanSDRangeMeanSDRangeMeanSDRange

Age(y:m)13:12:4(8:1-16:8)12:02:2(6:6-14:9)7:9*0:4(7:5-8:4)

VM A7:12:3(3:7-13:0)7:92:0(4:8-11:1)8:61:6(5:7-11:6)

BPVSstandardizedscoreAVISC-R VIQ58.714.2(40-92)67.311.0(46-85)

106.0*13.3(78-132)

Autism

M LD

Normal

25

2621

V


4/6

876 F. G. E. HAPPETable 3Frequency Data for Individual Illusions, Percentage of Subjects w ho Succum bed to Each IllusionGroupAutismTwo-dimensionalThree-dimensionalM LDTwo-dimensional

Three-dimensionalNormalTwo-dimensionalThree-dimensional

Ponzo28.024.034.619.252.428.6

Kanisza8.012.0

42.342.357.161.9

Titchener32.028.061.534.671.423.8

Muller-Lyer Hering88.060.088.530.880.928.6

40.024.069.246.276.233.3

Poggendorf28.036.057.726.961.938.1

three-dimensional condition the autism and controlgroups did not differ (F(2,69) = .40, p= .67). On multi-ple regression ANOVA there was a significant group bycondition interaction (F(2,69) = 9.95,p= .000).In terms of frequency data, only seven of the 25subjects with autism (i.e. 28%) succumbed to three ormore of the six two-dimensional illusions, comparedwith 20 (77%) of the MLD and 20 (95.2%) of the youngnormal controls. For the three-dimensional illusions, thecomparable frequencies were seven subjects with autism(28%), eight (31%) subjects with MLD and nine (42.9%)young normal subjects succumbing to half or more of theillusions. Most subjects in the control groups were aidedby the disembedded three-dimensional condition inmaking accurate judgements, i.e. most subjects suc-cumbed to fewer illusions in the three-dimensional formthan in the two-dimensional form. This was not true forthe subjects with autism. While less than half (44%) ofthe autism group was aided by the three-dimensionaldisembedded condition, almost all (95%) the normalcontrols and most (73%) of the MLD group were moreaccurate in the three-dimensional than the two-dimen-sional condition. Chi-square analysis showed that theautism group differed significantly in this respect fromboth control groups (autism vs normal, x^= 13.67,p .001; autism vs MLD, x^=4.44,p .05).

The pattern of results for the individual illusions isshown in Table 3. Of the six illusions, two showedsomewh at different results. The M uller-Lyer figure stoodout from the other illusions as the most 'persuasive' toall groups. Eighty-one per cent of the normal children,88 % of MLD subjects and, interestingly, 88% of theautism group succumbed to this illusion, erroneouslyjudgin g the lines to be of different lengths. This was theonly illusion which 'tricked' a significant proportion ofthe autism group. The other illusion w hich did not followthe pattern of result reported above, was the Kaniszatriangle which differed in being relatively unaffected bythe three-dimensional 'disembedded' condition (i.e. asmany errors were made on the three-dimensional versionof the triangle as on the two-dimensional version).

DiscussionThe present study was a first exploration of illusory

accurate judgements of illusory figures and less osuccumbed to the typical mis-judgements. That tsuccess on this task was due to failure to integrateinduced lines and the inducing context, is indicatedthe results from the three-dimensional condition. Wthe induced lines were artificially disembedded frominducing context, through colour and depth, consubjects performed like subjects with autism, andgroup differences disappeared. This effect is akinShah and Frith's (1993) finding with the block detask, where pre-segmenting the designs brought conperformance to the level of the autism group's succbut did not aid the subjects with autism. The theory people with autism have weak central coherence preand explains the present finding, and it is not appathat any other current theory of the disorder (e.g. defin executive function, theory of mind) can reaaccount for these results.Differences between the visual illusions used suggest that weak coherence may provide a gexplanation for the success of the autism group. only illusion to which a significant proportion ofautism group succumbed was the Muller-Lyer. illusion cannot easily be split into induced figure inducing context, since the arrow heads in the verused are connected to the judged lines and form parthe same 'object'. It is not clear exactly why percepof the Kan isza triang le was so little affected by the thdimensional condition, but it is worth noting that illusion is unlike the others used in that it indperception of illusory contours.The present findings are surprising in so far asvisual illusions appear to require integration at very levels, and it seemed unlikely a priori that subwould show w eak coherence at this level. It wilcourse, be necessary to replicate these results with osamples of subjects and different stimuli. More sopticated techniques are available for estimating magnitude of illusory effects (e.g. asking subjectadjust lines to match lengths), but it is impressive even using quite gross judgements and very simmethods, significant group differences could be founsubjects with autism do fail to integrate visual infortion at this low level, it may be necessary to think a


5/6

WE K CENTR L COHERENCE T LOW LEVELS 87 7Acknowledgements^The data reported in this paper were

References Diagnostic andstatistical manual of mental disorders (DSM-III-R) (3rd ed.,rev.). Washington, DC: American Psychiatric Association. Diagnostic andstatistical manual of mental disorders (DSM-IV) (4th ed.).Washington, DC: American Psychiatric Association.n, L. M., Dunn, L. M., Whetton, C , & Pintilie, D. (1982).British Picture Vocabulary Scale. Windsor, U.K.: NFER-Nelson. Autism: explaining the enigma. Oxford:Blackwell Scientific Publications.children. In D. J. Cohen, A. Donnellan, & R. Paul (Eds.),Hand book of autism and pervasive developm ental disorders(pp. 85-102). New York: John Wiley.mind' . Cognition, 50, 115-132reading for sound in autistic and dyslexic children. Journalof Developmental Psychology, 1, 329-342review of selected literature. Journal of OrthomolecularPsychiatry, 13, 144-175

Gregory, R. L. (1967). Eye and brain. New York: WorlUniversity Library.Happe, F. G. E. (1994a). Annotation: Psychological theories oautism. Journal of Child Psychology and Psychiatry, 35215-229Happe, F. (1994b).Autism: an introduction to psycho logicatheory. Lx)ndon: UCL Press.Happe, F. G. E. (1994c). Wechsler IQ profile and theory omind in autism: a research note. Journal of ChildPsychology and Psychiatry, 35, 1461-1471Happe, F. (in press). Central coherence and theory of mind inautism: reading homographs in context. British Journal oDevelopmental Psychology.Kanner, L. (1943). Autistic disturbances of affective contactNervous Child 2, 217-250Kanner, L. & Eisenberg, L. (1956). Early infantile autism1943-1955. Am erican Journal of Orthopsychiatry, 26, 5 565Rimland, B., & Hill, A. L. (1984). Idiot savants. In Wortis, J(Ed.), Mental retardation and developmental disabilitiesVol. 1 (pp. 155-169). New York: Plenum Press.Shah, A. & Frith, U. (1983). An islet of ability in autistichildren: a research note. Journal of Child Psychology andPsychiatry, 24, 613-620Shah, A. & Frith, U. (1993). Why do autistic individuals showsuperior performance on the Block Design task? Journal oChild Psychology and Psychiatry, 34, 1351-1364Wechsler, D. (1974). Wechsler Intelligence Scale for ChildrenRevised. New York: The Psychological CorporatioWilliams, D. (1992). Nobody nowhere. London: Doubleday.White, B. B. & White, M. S. (1987). Autism from the insideMedical Hypotheses, 24, 223-229

Accepted m anuscript received 29 February 199


6/6

visual illusions autism

Documents