Neuropsychologia 44 (2006) 161–170 A deficit in discriminating gaze direction in a case with right superior temporal gyrus lesion Tomoko Akiyama ∗ , Motoichiro Kato, Taro Muramatsu, Fumie Saito, Ryoko Nakachi, Haruo Kashima Komagino Hospital, 273 Uratakao-cho Hachioji City, Tokyo 193-8505, Japan Received 25 August 2004; received in revised form 22 April 2005; accepted 13 May 2005 Available online 6 July 2005 Abstract The superior temporal sulcus (STS) region is well recognized as being heavily involved in detecting and discriminating gaze. Lesions confined to this area are quite rare in humans, and so the research has mainly depended on animal studies and functional neuroimaging in normal human subjects. We report one such rare case, a 54-year-old Japanese female with a possible congenital s anomaly who, after a cerebral hemorrhage, demonstrated a lesion almost completely confined to the entire right superior temporal gyrus (STG). In the subacute phase, the patient showed evidence of left hemispatial neglect, from which she gradually recovered. In the chronic phase, she showed a puzzling difficulty in obtaining eye-contact. We have conducted, in conjunction with conventional neuropsychological evaluations, experimental assessment of her ability in gaze cognition. Her performance on neuropsychological testing demonstrated no compromise in intellect, memory, or language skills, and a close-to-full recovery from neglect. On gaze cognition experiments, she was repeatedly shown to perceive left gaze as straight, and to a lesser degree, straight gaze as right. We suggest that the function of the STG in detecting gaze, together with the directional information it receives from earlier visual areas, may be associated, when damaged, with this deficit in detecting contra-directional gaze. We have demonstrated for the first time that a single circumscribed lesion to the STG results in both gaze processing deficit and concurrent aberrant gaze behavior of the victim herself, implicating a mechanism within the STG as an interface between gaze of others and gaze of self. © 2005 Elsevier Ltd. All rights reserved. Keywords: Superior temporal sulcus; Superior temporal gyrus; Gaze; Biological motion; Hemianopia 1. Introduction The superior temporal sulcus (STS) has attracted much interest due to its intriguing function as a detector of biological motion. Since the 1980s, single-cell studies in monkey brains have shown that a number of neurons in the STS region show preference in the direction of head and gaze when viewing faces of conspecifics and humans, so that some neurons respond maximally to straight gaze while others respond more to averted gaze (Perrett, Hietanen, Oram, & Benson, 1992). Investigations of STS-lesioned monkeys showed disrupted gaze discrimination after lesioning (Campbell, Heywood, Cowey, Regard, & Landis, ∗ Corresponding author. Tel.: +81 426 63 2222; fax: +81 426 63 3286. E-mail address: tee-i@mxv.mesh.ne.jp (T. Akiyama). 0028-3932/$ – see front matter © 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuropsychologia.2005.05.018 1990; Eacott, Heywood, Gross, & Cowey, 1993), more or less supporting the findings from single-cell studies. Research of the STS in humans has found a homologous function; gaze as well as head direction, movement of mouth and hands, body gestures, geometric figures simulating biological motion have all been shown to activate the STS in functional imaging studies and electrophysiological studies (see Allison, Puce, & McCarthy, 2000; Puce & Perrett, 2003 for review). Thus, the function of the STS in humans can be summarized as the detection of biological motion, which composes a major contribution to social perception. Gaze, in particular, is undoubtedly special, perhaps without a match in its significance as a biological signal. There are suggestions for the innateness of gaze cognition, in that newborns show a preference for faces with eyes open versus eyes closed (Batki, Baron-Cohen, Wheelwright, Connellan, 162 T. Akiyama et al. / Neuropsychologia 44 (2006) 161–170 & Ahluwalia, 2000), and that infants as early as 10 weeks of age follow the gaze of others (Hood, Willen, & Driver, 1998). Moreover, children as young as 4–5 years of age show much more organized ERPs (N190) to eyes compared to full faces (Taylor, Edmonds, McCarthy, & Allison, 2001a), implicating a system within the human brain which is dedicated, at least in part, to gaze cognition. Such a system has been suggested to be localized in the posterior region of the STS in question, particularly in the right hemisphere: In a PET study, faces with visible gaze compared with obscured gaze, was shown to differentially activate the STS (Wicker, Michel, Henaff, & Decety, 1998). Gaze processing, such as perception of apparent eye motion (Puce, Allison, Bentin, Gore, & McCarthy, 1998) and detection of gaze direction (Hoffman & Haxby, 2000), has also been demonstrated to elicit significant STS activation in fMRI studies. ERP studies have shown that N170 from the posterior temporal region is also sensitive to gaze (Puce, Smith, & Allison, 2000; Taylor, Itier, Allison, & Edmonds, 2001b). Intriguing as it may be, however, the function of the STS in humans remains tentative in that most studies are based on functional findings of the intact brain. Neuropsychological studies of human cases with circumscribed STS lesions might be more conclusive, and would afford a wealth of insight into these issues. Since left STS damage would leave most humans aphasic with little understanding of any verbal input, study of a patient with right STS damage is likely to be more informative. To date, and to our knowledge, such a neuropsychological case has never been reported. In this paper, we report one such case, who, after a cerebrovascular accident, presented a very rare circumscribed lesion almost confined to the entire right superior temporal gyrus (STG), which constitutes the upper bank of the STS. The rarity of this lesion might have partly been due to a coexisting cerebrovascular anomaly in this patient. One of the most striking features of the sequelae in this patient was the inability to engage in mutual eye-contact. Given the functional data suggesting STS involvement in gaze cognition, a quantitative evaluation of the patient’s gaze processing would offer the first neuropsychological data for the STS and its suggested role in gaze cognition. We have therefore conducted, in addition to conventional neuropsychological assessments, some behavioral experiments in gaze processing which revealed a unique impairment in gaze cognition. high density area in the right temporal lobe with a midline shift to the left, consistent with an acute cerebral hemorrhage. Cranial angiography revealed absence of the right internal carotid artery and depicted the left internal carotid artery feeding into the right middle cerebral artery region, pointing out a possible anomaly of the cranial arteries. Surgery was performed to remove the hematoma on the day of onset. After surgery, she quickly regained consciousness and her hemiparesis resolved within a course of a month. However, shortly after regaining consciousness, she complained of leftsided visual impairment. Upon visual field perimetry, a left homonymous hemianopia became evident. She also initially showed signs of left hemispatial neglect unexplainable with mere hemianopia, from which she gradually recovered within a course of a year. A puzzling behavior caught our attention after several sessions. When seated face to face with M.J., she often revealed a difficulty in obtaining eye-contact, with a tendency for M.J.’s gaze to drift to her contralateral, left. Her family confirmed that she has had no such difficulty prior to her cerebrovascular accident. On the other hand, she had no trouble fixating on objects when explicitly instructed to, as in the case where she was asked to look at the eyes of the observer. Thus, the drift in her gaze to the left was seen at rest, or during casual conversation, when M.J. was not keenly attending to any object. The MRI in the chronic, stable phase demonstrated a rare, circumscribed lesion in the right STG, which slightly extended into the white matter underlying the gyrus (Fig. 1). The hemianopia can be explained by the lesion extension to the posterior horn of the lateral ventricle, thereby affecting the optic radiation. Also shown is a moderate atrophy of the right medial temporal structures compared to the intact left structures. 3. Neuropsychological assessment Assessment was performed with conventional neuropsychological tests on two occasions (Table 1). The first assessment was administered 2–4 months after the onset of the hemorrhage when M.J. still had dense hemispatial neglect. The second assessment was delivered a year after the first, when the neglect had remitted. Between these two assessments, M.J. visited the outpatient department twice a month for non-structured rehabilitation. 3.1. Results 2. Case description The patient, M.J. is a right-handed, Japanese female with high school education who has been married for over 30 years and has successfully reared two children. She has no prior neurological or psychiatric history. At the time of onset, she was 54 years old. She presented with headache, nausea, slight left hemiparesis, and clouded consciousness. She was transferred to a nearby hospital, where her cranial CT revealed a In the first assessment, M.J. performed within the normal range for intelligence, but showed a large discrepancy between verbal and visual material. Hemispatial neglect may explain most of the problem M.J. had on visual material, as can be seen in her performance for cancellation tests and line bisections. Detailed results are shown in Table 1. In the second assessment, M.J. showed much improvement in many of the tests. The performance IQ in the WAIS-R T. Akiyama et al. / Neuropsychologia 44 (2006) 161–170 163 Fig. 1. MRI scan 2 years post onset. A rare lesion almost completely circumscribed to the entire STG, which is indicated by the arrows, is shown in the (a) axial, (b) sagittal, and (c) coronal slices. A moderate atrophy of the right medial temporal lobe can also be observed in the coronal slice. was much better although there still remained a fair discrepancy between verbal IQ. Visual memory in the WMS-R also improved. On the cancellation test she omitted less. Her line bisections no longer deviated to the right, but now showed a minimal deviation to the left, which is not significantly deviant from pseudoneglect often demonstrated in normal subjects (Jewell & McCourt, 2000). Hemispatial neglect was assessed using the behavioral inattention test (BIT), a battery which has been validated and standardized in patients with stroke (Wilson, Cockburn, & Halligan, 1987a, 1987b). Her BIT score was well within the normal range, further indicating her recovery from the neglect syndrome. The performances on the Rey auditory verbal learning test were superior. Overall, M.J.’s neuropsychological assessment showed that although she initially demonstrated signs of left hemispa- tial neglect, it had nearly diminished within a course of a year. With this improvement, she presently shows no compromise in intellect, memory, or language skills. 4. Experiment 1: gaze direction discrimination in digitized photographs This and the following experimental investigations were carried out from the time of the second neuropsychological assessment, for a period of about a year. M.J. gave informed consent to participation in all experimental investigations. This study was approved by the ethical committee at our institutions. Note that all indications of ‘right’ and ‘left’ will refer to the direction from M.J.’s viewpoint. Note also that all T. Akiyama et al. / Neuropsychologia 44 (2006) 161–170 164 Table 1 Neuropsychological scores Test First assessment Wechsler adult intelligence scale—revised Verbal IQ 108 Performance IQ 87 Full IQ 99 Raven’s coloured 32/36 progressive matrices Wechsler memory scale—revised Verbal memory 110 Visual memory 70 General memory 97 Delayed recall 93 Attention/concentration 98 Rey complex figure test Copy Recall 36/36 20.5/36 Second assessment 117 101 111 34/36 115 106 114 107 101 33/36 15/36 Rey auditory verbal learning test Immediate recall –a Post-interference recall – 10, 12, 14, 14, 15/15 15/15 Cancellation test Numeral 3 113/114 (99%), 104 s 108/114 (95%), 133 s Japanese character ‘ka’ 99/114 (87%), 141 s Line bisection test (deviation as % total length) Midsagittal presentation 20.5% to right Right hemispace presentation 18.5% to right Left hemispace presentation 25.0% to right Behavioral inattention test Conventional subsets Behavioral subsets – – 103/114 (90%), 130 s 4.5% to left 5.0% to left 5.5% to left 144/146 80/81 a Not administered. materials were presented in her midsagittal plane, and that she was free to move her head and eyes during the tasks, unless stated otherwise. A 77-year-old female patient with left homonymous hemianopia due to right thalamic hemorrhage (left hemianopic control, LH), a 43-year-old male with right homonymous hemianopia after left occipital lobe hemorrhage (right hemianopic control A, RH-A), and a 51-year-old male with right homonymous hemianopia due to traumatic brain injury of the optic radiation (right hemianopic control B, RH-B) served as hemianopic controls in all the experiments. Ten healthy adults (five males, five females, mean age 37.5) also participated in Experiments 1 and 3, and five from this group in Experiment 2, as normal subjects. right/left inversion. A total of six photographs, i.e. S, R, L, S-inv, R-inv, and L-inv were obtained. The photographs were trimmed so that only the eye region, including the eyebrows, was shown (70 mm × 32 mm, Fig. 2a). In each photograph, the width of one eye measured approximately 16 mm, and the diameter of the iris 6 mm. Fifteen copies of each six photographs were made and placed in a pseudorandom order to make a set of 90 consecutive photographs of the eye region with three different gaze directions, and were shown singly on a 248 mm × 188 mm PC screen. The viewing distance was approximately 400 mm. M.J. was asked to report the direction of the gaze as right, left, or straight from her viewpoint. She was given unlimited time to respond, but all her responses were prompt, rarely taking more than 2 s. 4.2. Results M.J. discriminated the gaze direction correctly in 68 out of 90 photographs, giving her an error rate of 24.4%, which is substantially higher than that of the hemianopic controls (LH 14.4%, RH-A 1.1%, RH-B 7.8%), and normal controls (3.3%). Of the three directions, left gaze had the highest error rate for M.J., 40% of which were incorrectly perceived as straight gaze (Fig. 2b). As a contrast, LH misperceived right gaze as straight (76.9% of the errors were misperception of right gaze as straight, the remaining 23.1% were misperception of straight gaze as left), and RH-B made random errors without any directionality. A χ2 analysis was conducted to assess the differences in the distribution of error responses. M.J.’s error ratio was significantly higher than RH-A, RH-B and normal controls (p < 0.01). No significant difference was observed between LH and normal controls (p = 0.24). Although not reaching significance, M.J. showed a tendency to have more errors than LH (p = 0.09) in addition to the opposite bias in the direction of errors (M.J., rightward; LH, leftward). 5. Experiment 2: mutual gaze detection Given the suggestion that mutual gaze might be significant in its own right (George, Driver, & Dolan, 2001; Kampe, Frith, Dolan, & Frith, 2001; Kawashima et al., 1999), we wanted to test whether M.J. had preserved cognition of this fundamental aspect of gaze—he detection of mutual gaze. We realized that neither photographs nor videotapes would be a true equivalent of a real-life mutual gaze, so we designed this experiment with real-life gaze in a more natural setting. 5.1. Method 4.1. Method Three digitized photographs of a female looking straight at the camera (S), 8◦ to the right (R) and to the left (L) of the camera were taken. To control for any asymmetry of the face or the eyes, each picture was then reproduced with a An experimenter well acquainted with M.J. stood face to face with her, one meter away, with M.J. standing against a wall. On the wall behind M.J. and to both sides were 11 markers at the same height as M.J.’s eyes. The five markers closest to M.J. were 60 mm apart, and the remaining six T. Akiyama et al. / Neuropsychologia 44 (2006) 161–170 165 Fig. 2. (a) Examples of photographs used in Experiment 1. (b) Results of M.J.’s performance in Experiment 1. Gaze direction was discriminated in 90 digitized photographs (left gaze, 30; straight gaze, 30; right gaze, 30). M.J.’s performance is shown in relation to each gaze direction. were 90 mm apart, comprising a gaze angle of approximately 36◦ for the farthest marker. The experimenter pursued each marker from either the extreme right or the left, alternating the starting point with each trial. Each target was viewed for approximately 1 s. When the experimenter’s gaze reached the marker right next to M.J.’s face, the next marker was her nasal bridge, followed by the markers on the other side. M.J. was asked to look straight ahead at the experimenter’s eyes during these trials, and to inform the experimenter when she felt their eyes met. M.J.’s eye-position was monitored by the observation of an independent experimenter to assure that she maintained straight gaze. The experiment consisted of 12 trials, and the closest marker being looked at when M.J. reported mutual gaze detection was recorded. The distance of these markers from M.J.’s nasal bridge was determined, then calculated into degrees of deviation from her nose. trials beginning from the right. For RH-A and RH-B, and for all five normal control subjects, the average deviations were 0◦ and 0◦ , respectively, for either direction. 5.2. Results Rectangular schematic eyes consisted of rectangular diagrams in pairs. A black rectangle (6 mm × 10 mm) which was made to resemble the iris, was placed within a white rectangle (16 mm × 10 mm) resembling the sclera. The horizontal side of the black rectangle had a length matching the diameter of the iris in Experiment 1, and the white rectangle had a horizontal side matching the width of the eyes in Experiment 1. There were three orientations of the black rectangle in relation to the white; in the center (C), or slightly deviated to the right (R) or the left (L) of the center. The deviation was approximately 6.5% off the center relative to the entire horizontal length, which in effect was comparable to the right/left deviation of the gaze in Experiment 1. Two such identical diagrams were shown in pairs in a horizontal array (Fig. 3a). Throughout the experiment, M.J. did not show any difficulty following the instruction to look straight ahead. In the six trials where the pursuit began at M.J.’s extreme left, she was shown to perceive gaze looking slightly to the left of her face as mutual gaze, with an average deviation of 11.1◦ to the left. In another words, gaze slightly deviated to M.J.’s left was perceived as mutual, showing the same directionality in gaze misperception as seen in Experiment 1. On the other hand, M.J. was much more precise in detecting mutual gaze in the six trials where the pursuit began at her extreme right, with an average deviation of 0◦ . The average deviations for LH were 0◦ for trials beginning from the left, and 7.0◦ to the right for 6. Experiment 3: direction discrimination in schematic eyes To investigate whether the gaze discrimination impairment seen in Experiments 1 and 2 were gaze-specific, and not just a reflection of the residual spatial inattention per se, two types of schematic eyes (rectangular and elliptical) were designed and used as control tasks in direction discrimination. 6.1. Method 166 T. Akiyama et al. / Neuropsychologia 44 (2006) 161–170 Fig. 3. Examples of (a) rectangular and (b) elliptical schematic eyes used in Experiment 3. Results of M.J.’s performance in (c) rectangular and (d) elliptical schematic eyes of Experiment 3. The direction of the schematic ‘iris’ was discriminated in 45 schematic eyes (left-oriented schema, 15; straight-oriented schema, 15; right-oriented schema, 15) for both configurations. M.J.’s performance is shown in relation to each ‘gaze’ orientation. Fifteen copies of each orientation were made and placed in a pseudorandom order to make a set of 45 consecutive diagrams with three different orientations of the black ‘iris’, and were shown on the same PC screen. Elliptical schematic eyes consisted of diagrams in pairs again, but this time with black circles (diameter: 6 mm) for the iris and white ellipses (axes: 16 mm × 10 mm) for the sclera. The size of the circles and ellipses were made to match the size of the iris and the eye in Experiment 1. The schematic iris was oriented either in the center (C) of the schematic sclera, or slightly deviated to the right (R) or the left (L) of the center, with comparable deviation as in Experiment 1 (approximately 6.5% off center), and were shown in pairs (Fig. 3b). These were also made into a set of 45 diagrams, and were shown on the PC screen. The viewing distance was approximately 400 mm. M.J. was asked, first for the rectangles, then for the ellipses, to report the orientation of the black figures in relation to the white, with right, left, or center from her viewpoint. Again, she was given unlimited time, but her responses were prompt. One thing of note is that the resemblance of these diagrams to eyes was never explicitly mentioned by the experimenter. which contrasts sharply with the good performance for both control groups (LH 0%, RH-A 8.9%, RH-B 0%, normal controls 0.4%). A χ2 analysis demonstrated that M.J.’s error ratio was significantly higher than LH, RH-A, RH-B, and normal controls (p < 0.01). In respect to each orientation in M.J.’s performance, R ellipses were never misjudged, compared to the surprisingly high error rates of 66.7 and indeed, 100% for the C and L ellipses, respectively (Fig. 3d). Most strikingly, all the errors that M.J. made reflected a directional bias to the right throughout both configurations; L diagrams had a strong tendency to be erroneously perceived as C, and C diagrams as R. RH-A also made systematic errors, albeit slighter, in the opposite direction; R diagrams were misjudged as C, and C diagrams as L. However, the error rates were identical between the two configurations. On the other hand, M.J. showed a much higher error rate for the ellipses compared to the rectangles, demonstrating a relative specificity of her impairment to stimuli which more closely resemble the eyes. It is worth mentioning that it was after the 6.2. Results For the rectangles, M.J. discriminated the orientation of the black ‘iris’ correctly in 39 out of 45, giving her an error rate of 13.3%, which was higher than that of the hemianopic controls (LH 4.4%, RH-A 8.9%, RH-B 0%), and substantially higher than that of normal controls (0.4%). In respect to each orientation in M.J.’s performance, none of the rightoriented rectangles were misjudged, but 33.3% and 6.7% of the centrally oriented (C), and left-oriented (L) rectangles were misjudged, respectively (Fig. 3c). For the ellipses, M.J. discriminated the orientation correctly in 20 out of 45, giving her an error rate of 55.6%, Fig. 4. Summary of the results for all participants in Experiment 1 (discriminating gaze direction from photographs of eyes) and Experiment 3 (discriminating ‘gaze’ direction from rectangular and elliptical schematic eyes) is shown as error rates for each stimulus type. x-Axis, type of stimuli tested; y-axis, error rate in percentage. LH, left hemianopia; RH-A and B, right hemianopia; NC, normal control. T. Akiyama et al. / Neuropsychologia 44 (2006) 161–170 trial with the rectangles, and at the very beginning of the trial with the elliptical eyes that M.J. spontaneously became aware of the resemblance of the diagrams to eyes. Possibly with this awareness, the error rate of L and C diagrams significantly increased from 20.0% for the rectangles, to 83.3% for the ellipses. A conceptual rephrasing of this finding might be that when the diagrams were perceived, not as geometric diagrams but as schematic eyes, more than 80% of the left ‘gaze’ and straight ‘gaze’ were misjudged with a heavy directional bias to the right. A summary of the results for Experiments 1 and 3 is shown in Fig. 4. 7. Discussion M.J., who after a near-complete damage of the right STG in her fifth, heretofore uneventful decade, became longstandingly hemianopic, but recovered gradually from her initial left hemispatial neglect with some subtle residual impairments that were no longer detectable by conventional neuropsychological assessment. The most puzzling of these impairments was a difficulty in obtaining eye contact. Taken together the implication that the STS (the upper bank of which comprises the STG) is involved heavily in gaze detection, we were prompted to evaluate M.J.’s gaze cognition in detail. As has been shown, a unique directional impairment in gaze discrimination was demonstrated; M.J. was repeatedly shown to perceive left gaze as straight, and to a lesser degree, straight gaze as right in photographs (Experiment 1), in real encounters (Experiment 2), and in schematic eyes (Experiment 3). To the best of our knowledge, such impairment has never been reported in association with human brain damage. For the first time, then, has the repeated implication of STS involvement in gaze processing been supported neuropsychologically in humans. 7.1. The directional property of gaze It actually does not come as too much of a surprise, given animal lesioning data (Campbell et al., 1990) and human functional data (Hoffman & Haxby, 2000; Pelphrey, Singerman, Allison, & McCarthy, 2003; Puce et al., 1998, 2000; Taylor et al., 2001b; Wicker et al., 1998), that a human lesion of the STS should also impair gaze cognition, as seen in this case. What came across as surprising, however, was the finding of a directionality (i.e., the consistent rightward bias) in M.J.’s perception of gaze, specifically when she currently demonstrates minimal if any signs of neglect in standardized testing. The impairment could be termed as a deficit in discriminating contralesional gaze direction. As such an impairment is quite novel, we believe it deserves further discussion. There is growing body of research which elegantly demonstrates the automatic direction-cueing role of gaze. Friesen and Kingstone (1998) showed, in normal subjects, that targets 167 shown with schematic faces gazing in their direction elicited shorter detection time compared to targets with schema gazing in the opposite direction. Even more compelling are the findings that suggest that gaze, when compared to other nonbiological directional cues, has a distinct directional valence of its own. Gaze, but not arrows, has been demonstrated to elicit gaze-following behavior in adults (Ricciardelli, Bricolo, Aglioti, & Chelazzi, 2002). Zorzi, Mapelli, Rusconi, and Umilta (2003) have found that gaze direction has an effect independent from that cued by an egocentricallydefined directional signal, in behavioral tasks discriminating direction. In an fMRI study, Hooker et al. (2003) have demonstrated that direction-informing gaze differentially activates the STS, compared to arrows, non-directional eye motion, and even arrows superimposed on a photograph with uninformative gaze. Additional support for the involvement of the STS in detecting direction from gaze is offered by Pelphrey et al. (2003) who demonstrated a difference in MR signal waveform in the STS, between conditions where the schematic gaze oriented correctly or incorrectly in the direction of the target. These findings suggest three important points. Firstly, gaze has a definitive direction-cueing property. Secondly, gaze, in contrast to other directional cues, is in the position to differentially attract the gaze and attention of the viewer. Lastly, this directional information derived from gaze seems to be processed, at least to some extent, within the STS as shown in the last two fMRI studies. How is the STS in the position to process such directional information? Two lines of evidence might be relevant to this question. A region termed the medial temporal/medial superior temporal cortex (MT/MST) in the monkey brain, which lies posterior to, and subsequently earlier in the visual stream than the STS, is recognized as a visual area tuned selectively to the direction of motion (see Wurtz, Yamasaki, Duffy, & Roy, 1990 for review). In humans, a group of unilateral braindamaged subjects whose lesion overlapped in Brodmann area 19/37 were demonstrated to have a directional deficit in motion perception (Barton, Sharpe, & Raymond, 1995, 1996; Vaina, Cowey, Eskew, LeMay, & Kemper, 2001a) and smooth pursuit (Barton et al., 1996). This area at the temporo-parietooccipital (TPO) junction, also activated in functional imaging studies tapping motion perception processing (Tootell et al., 1995; Watson et al., 1993), has since been considered the human homologue of MT/MST. The direction-sensitive information from the MT/MST is known to terminate at the superior temporal polysensory area in the STS of the monkey brain, which is precisely where the processing of biological motion, gaze being one of them, is implicated in humans (Bonda, Petrides, Ostry, & Evans, 1996; Vaina, Solomon, Chowdhury, Sinha, & Belliveau, 2001b). The second line of evidence comes from a report by Karnath, Ferber, and Himmelbach (2001), studying 25 human patients with hemispatial neglect, which revealed that the center of maximal lesion overlap among these patients was in the STG. This finding might lend additional support that there is indeed some directional processing in this gyrus. To summarize, the 168 T. Akiyama et al. / Neuropsychologia 44 (2006) 161–170 STS (or the STG) in question is well in the position to receive directional information pertaining to the biological motion it processes. Accordingly, we have found for the first time, in a right STG-damaged case, a contra-directional impairment in one of the most robust biological motions; gaze. It does not go beyond speculation, but this finding might imply that leftward gaze is represented in the right STS (and perhaps rightward gaze in the left STS). 7.2. Influence of hemianopia It has been shown, in hemianopics without neglect, that horizontal lines in the compromised visual field have a tendency to be overestimated, or perceived as longer than their true length, while lines in the intact field are underestimated, or perceived as shorter than their true length (Barton, Behrmann, & Black, 1998; Doricchi, Onida, & Guariglia, 2002). A good illustration of this tendency is shown in line bisections of such subjects, where the subjective midpoints fall into the contralesional half of the line. Normal subjects are also shown, when viewing lines only in their right visual field, to have a tendency to bisect them centripetally, or to the left, of the actual center (Nielsen, Intriligator, & Barton, 1999). These consistent findings may offer a completely different explanation for M.J.’s directional gaze impairment; as a left hemianopic, she over-represents the left side of the sclera, resulting in rightward misrepresentation of the iris. However, we argue against this interpretation as the sole mechanism for M.J.’s gaze processing deficit under the following grounds. First of all, the left hemianopic control patient performed better than M.J. in all the experiments. More importantly, the error pattern that this control patient demonstrated in Experiments 1 and 2 was of a leftward bias, opposite to that of M.J. Although this control patient does not presently demonstrate robust left neglect, her performance on the BIT (Conventional subsets 133/146, Behavioral Subsets 74/81), though above the cutoff points, are worse than that of M.J. (144/146 and 80/81, respectively). This small difference might have affected this control patient’s performance; the leftward bias that she demonstrated might have perhaps been a reflection of residual left neglect, i.e., the left side of the sclera is neglected, resulting in leftward misrepresentation of the iris. On the other hand, this left hemianopia control showed little compromise in Experiment 3, demonstrating clearly that the poor performance thereof on M.J.’s part cannot be explained by left hemianopia, nor left neglect for that matter. It is indeed difficult to disentangle left hemianopia from left neglect in right hemisphere damage (Daini, Angelelli, Antonucci, Cappa, & Vallar, 2002; Walker, Findlay, Young, & Welch, 1991), and in this sense, right hemianopia might control for a purer hemianopic effect unaffected by right hemispheric spatial impairment. Turning, then, to the right hemianopic controls, both were essentially unimpaired in Experiments 1 and 2. It is true that in Experiment 3, one of the right hemianopic controls (RH-A) made substantial directional errors which might be accounted for by his hemianopia, but the error rates remained constant between rectangular and elliptical configurations. It is worth emphasizing that none of the hemianopic controls, left or right, showed the quadruple increase in errors that M.J. demonstrated for elliptical, compared to rectangular eyes (Fig. 4). Thus, a mechanism independent from hemianopia, and specific to gaze, is called for. We conclude that, as much as a contribution of hemianopia might seem likely, as seen in the comparable error rates of M.J. and RHA for the rectangular eyes, hemianopia also cannot explain M.J.’s core deficit in gaze processing. Instead, the involvement of the STG seems to have been essential in causing her impairment. 7.3. Your gaze, my gaze One final observation will be discussed here; M.J.’s difficulty in engaging in mutual eye-contact, with a tendency of her gaze to drift to her contralesional left. Patients with injuries to the unilateral TPO junction are known to be impaired in pursuing targets with their eyes (Morrow & Sharpe, 1990, 1993; Thurston, Leigh, Crawford, Thompson, & Kennard, 1988). It is quite ecological that this smooth pursuit center should lie so closely together with the gaze processing region; it is advantageous for the gaze-following behavior so automatic (Ricciardelli et al., 2002), so essential (Baron-Cohen, 1995), and perhaps innate (Hood et al., 1998) to humans. It might well be said that intact gaze cognition is essential in developing normal gaze behavior like eye-contact and gaze-following. For example, autistic children, whose lack of reciprocal gaze exchange is one of the most striking manifestations, have been shown to demonstrate aberrant gaze processing (Senju, Tojo, Dairoku, & Hasegawa, 2004; Senju, Yaguchi, Tojo, & Hasegawa, 2003). Likewise, gaze processing impairment due to STG damage, as seen in this case, could also be speculated to hamper the gaze behavior of the victim herself. Another potential interpretation might be that the leftward deviation in M.J.’s eye-position is in itself sufficient in explaining her rightward bias in gaze perception. We would like to point out, however, that M.J. was not observed to show any trouble fixating on the stimuli presented to her in Experiments 1 and 3, and was also able to maintain straight gaze when instructed to, in Experiment 2, as has been monitored by an independent observer. Additionally, had there been a mass effect of eye-position that the experimenters overlooked, the results for the rectangular and elliptical eyes in Experiment 3 should have been quite similar. Instead, a clear discrepancy is evident, favoring our hypothesis that the results seen in our experiments are a reflection of a deficit in gaze cognition, and not eye-position. Most importantly, that the gaze processing of others should be impaired confluent with the gaze behavior of self, as an aftermath of a single circumscribed brain lesion is quite intriguing. One interpretation would be that these two phenomena are independent from each other and thus unrelated; T. Akiyama et al. / Neuropsychologia 44 (2006) 161–170 the leftward drift in her eye-position might be accounted for, not by her gaze processing deficit, but for example, by compensation for her left hemianopia. However, we have noted no difficulty in eye-contact, nor robust drift in eye-position in the three hemianopic control subjects. A larger control group might clarify the relationship between hemianopia and poor eye-contact. The second, and perhaps the more attractive interpretation might be that the two phenomena are indeed associated. If we were to imagine a window-like grid between two people facing each other, mutual gaze would be aligned with the midsagittal window. In the case of M.J., on the other hand, gaze looking through a window to M.J.’s left would be misperceived as mutual, and in an attempt to engage in mutual eye-contact herself, she might also direct her gaze in the direction of that left window. If we were to assume this position that the two phenomena are associated, it would be implicative of a mechanism in the STG which serves as an interface between the gaze of others and the gaze of self in its response. This neuropsychological case might be the first to demonstrate a neural ground for the intuition (and the substantial behavioral data), that your gaze is intimately associated with mine. 7.4. Limitations The limitations in this study leave some questions unanswered. Among them are: (1) Why does M.J. perform better for photographic eyes than schematic elliptical eyes, and why are the hemianopic control subjects inconsistent as to the stimuli he/she performs better, photographic or schematic? Normal control data suggest that the photographs are the more difficult stimuli (Fig. 4). For M.J., however, the photographs were worse than rectangular eyes, but better than elliptical eyes. The performances for RH-A also discord from normal data. A larger study with more control subjects and diverse stimuli might offer the answers. (2) Would M.J. show preserved or impaired attention-cueing effect of gaze (for example, Friesen & Kingstone, 1998; Kingstone, Tipper, Ristic, & Ngan, 2004), and what would her eye-movement pattern be like in such a paradigm? Basing on the data we have demonstrated in this study, we speculate that left gaze would be less effective than right gaze in orienting M.J.’s attention. We are hoping to explore into these issues in the future. 8. Conclusions This study has demonstrated, in a case with a rare right STG damage, a novel deficit in discriminating contralesional gaze direction. We have suggested that the function of the STG in processing gaze, together with the directional information it derives from earlier visual areas like the MT/MST, may be associated, when damaged, with such a deficit. Concurrent with this deficit was an impaired ability to obtain eye-contact, implicating that the STG might also serve as an 169 interface between the gaze of others, and gaze of self. As left hemispheric damage to the same region would obviously leave patients heavily aphasic, this rare case might serve as a prototype of human STG damage. References Allison, T., Puce, A., & McCarthy, G. (2000). Social perception from visual cues: Role of the STS region. Trends in Cognitive Sciences, 4(7), 267–278. Baron-Cohen, S. (1995). Mindblindness: An essay on autism and theory of mind. Cambridge, MA: MIT Press. Barton, J. J., Behrmann, M., & Black, S. (1998). Ocular search during line bisection. The effects of hemi-neglect and hemianopia. Brain, 121(Pt 6), 1117–1131. Barton, J. J., Sharpe, J. A., & Raymond, J. E. (1995). Retinotopic and directional defects in motion discrimination in humans with cerebral lesions. Annals of Neurology, 37(5), 665–675. Barton, J. J., Sharpe, J. A., & Raymond, J. E. (1996). Directional defects in pursuit and motion perception in humans with unilateral cerebral lesions. Brain, 119(Pt 5), 1535–1550. Batki, A., Baron-Cohen, S., Wheelwright, S., Connellan, J., & Ahluwalia, J. (2000). Is there an innate gaze module? Evidence from human neonates. Infant Behavior and Development, 23, 223–229. Bonda, E., Petrides, M., Ostry, D., & Evans, A. (1996). Specific involvement of human parietal systems and the amygdala in the perception of biological motion. Journal of Neuroscience, 16(11), 3737–3744. Campbell, R., Heywood, C. A., Cowey, A., Regard, M., & Landis, T. (1990). Sensitivity to eye gaze in prosopagnosic patients and monkeys with superior temporal sulcus ablation. Neuropsychologia, 28(11), 1123–1142. Daini, R., Angelelli, P., Antonucci, G., Cappa, S. F., & Vallar, G. (2002). Exploring the syndrome of spatial unilateral neglect through an illusion of length. Experimental Brain Research, 144(2), 224–237. Doricchi, F., Onida, A., & Guariglia, P. (2002). Horizontal space misrepresentation in unilateral brain damage. II. Eye-head centered modulation of visual misrepresentation in hemianopia without neglect. Neuropsychologia, 40(8), 1118–1128. Eacott, M. J., Heywood, C. A., Gross, C. G., & Cowey, A. (1993). Visual discrimination impairments following lesions of the superior temporal sulcus are not specific for facial stimuli. Neuropsychologia, 31(6), 609–619. Friesen, C., & Kingstone, A. (1998). The eyes have it! Reflexive orienting is triggered by nonpredictive gaze. Psychonomic Bulletin and Review, 5(3), 490–495. George, N., Driver, J., & Dolan, R. J. (2001). Seen gaze-direction modulates fusiform activity and its coupling with other brain areas during face processing. Neuroimage, 13(6 Pt 1), 1102–1112. Grossman, E., Donnelly, M., Price, R., Pickens, D., Morgan, V., Neighbor, G., et al. (2000). Brain areas involved in perception of biological motion. Journal of Cognitive Neuroscience, 12(5), 711–720. Hoffman, E. A., & Haxby, J. V. (2000). Distinct representations of eye gaze and identity in the distributed human neural system for face perception. Nature Neuroscience, 3(1), 80–84. Hood, B. M., Willen, J. D., & Driver, J. (1998). Adult’s eyes trigger shifts of visual attention in human infants. Psychological Science, 9, 53–56. Hooker, C. I., Paller, K. A., Gitelman, D. R., Parrish, T. B., Mesulam, M. M., & Reber, P. J. (2003). Brain networks for analyzing eye gaze. Brain Research: Cognitive Brain Research, 17(2), 406–418. Jewell, G., & McCourt, M. E. (2000). Pseudoneglect: A review and meta-analysis of performance factors in line bisection tasks. Neuropsychologia, 38(1), 93–110. Kampe, K. K., Frith, C. D., Dolan, R. J., & Frith, U. (2001). Reward value of attractiveness and gaze. Nature, 413(6856), 589. 170 T. Akiyama et al. / Neuropsychologia 44 (2006) 161–170 Karnath, H. O., Ferber, S., & Himmelbach, M. (2001). Spatial awareness is a function of the temporal not the posterior parietal lobe. Nature, 411(6840), 950–953. Kawashima, R., Sugiura, M., Kato, T., Nakamura, A., Hatano, K., Ito, K., et al. (1999). The human amygdala plays an important role in gaze monitoring. A PET study. Brain, 122(Pt 4), 779–783. Kingstone, A., Tipper, C., Ristic, J., & Ngan, E. (2004). The eyes have it!: An fMRI investigation. Brain and Cognition, 55(2), 269–271. Morrow, M. J., & Sharpe, J. A. (1990). Cerebral hemispheric localization of smooth pursuit asymmetry. Neurology, 40(2), 284–292. Morrow, M. J., & Sharpe, J. A. (1993). Retinotopic and directional deficits of smooth pursuit initiation after posterior cerebral hemispheric lesions. Neurology, 43(3 Pt 1), 595–603. Nielsen, K. E., Intriligator, J., & Barton, J. J. (1999). Spatial representation in the normal visual field: A study of hemifield line bisection. Neuropsychologia, 37(3), 267–277. Pelphrey, K. A., Singerman, J. D., Allison, T., & McCarthy, G. (2003). Brain activation evoked by perception of gaze shifts: The influence of context. Neuropsychologia, 41(2), 156–170. Perrett, D. I., Hietanen, J. K., Oram, M. W., & Benson, P. J. (1992). Organization and functions of cells responsive to faces in the temporal cortex. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences, 335(1273), 23–30. Puce, A., Allison, T., Bentin, S., Gore, J. C., & McCarthy, G. (1998). Temporal cortex activation in humans viewing eye and mouth movements. Journal of Neuroscience, 18(6), 2188–2199. Puce, A., & Perrett, D. (2003). Electrophysiology and brain imaging of biological motion. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences, 358(1431), 435–445. Puce, A., Smith, A., & Allison, T. (2000). ERPs evoked by viewing facial movements. Cognitive Neuropsychology, 17, 221–239. Ricciardelli, P., Bricolo, E., Aglioti, S. M., & Chelazzi, L. (2002). My eyes want to look where your eyes are looking: Exploring the tendency to imitate another individual’s gaze. Neuroreport, 13(17), 2259–2264. Senju, A., Tojo, Y., Dairoku, H., & Hasegawa, T. (2004). Reflexive orienting in response to eye gaze and an arrow in children with and without autism. Journal of Child Psychology and Psychiatry, 45(3), 445–458. Senju, A., Yaguchi, K., Tojo, Y., & Hasegawa, T. (2003). Eye contact does not facilitate detection in children with autism. Cognition, 89(1), B43–B51. Taylor, M. J., Edmonds, G. E., McCarthy, G., & Allison, T. (2001). Eyes first! Eye processing develops before face processing in children. Neuroreport, 12(8), 1671–1676. Taylor, M. J., Itier, R. J., Allison, T., & Edmonds, G. E. (2001). Direction of gaze effects on early face processing: Eyes-only versus full faces. Brain Research: Cognitive Brain Research, 10(3), 333– 340. Thurston, S. E., Leigh, R. J., Crawford, T., Thompson, A., & Kennard, C. (1988). Two distinct deficits of visual tracking caused by unilateral lesions of cerebral cortex in humans. Annals of Neurology, 23(3), 266–273. Tootell, R. B., Reppas, J. B., Kwong, K. K., Malach, R., Born, R. T., Brady, T. J., et al. (1995). Functional analysis of human MT and related visual cortical areas using magnetic resonance imaging. Journal of Neuroscience, 15(4), 3215–3230. Vaina, L. M., Cowey, A., Eskew, R. T., Jr., LeMay, M., & Kemper, T. (2001). Regional cerebral correlates of global motion perception: Evidence from unilateral cerebral brain damage. Brain, 124(Pt 2), 310–321. Vaina, L. M., Solomon, J., Chowdhury, S., Sinha, P., & Belliveau, J. W. (2001). Functional neuroanatomy of biological motion perception in humans. Proceedings of the National Academy of Sciences of the United States of America, 98(20), 11656–11661. Walker, R., Findlay, J. M., Young, A. W., & Welch, J. (1991). Disentangling neglect and hemianopia. Neuropsychologia, 29(10), 1019– 1027. Watson, J. D., Myers, R., Frackowiak, R. S., Hajnal, J. V., Woods, R. P., Mazziotta, J. C., et al. (1993). Area V5 of the human brain: Evidence from a combined study using positron emission tomography and magnetic resonance imaging. Cerebral Cortex, 3(2), 79– 94. Wicker, B., Michel, F., Henaff, M. A., & Decety, J. (1998). Brain regions involved in the perception of gaze: A PET study. Neuroimage, 8(2), 221–227. Wilson, B., Cockburn, J., & Halligan, P. (1987a). Behavioural inattention test. Titchfield, Hampshire: Thames Valley Test Company. Wilson, B., Cockburn, J., & Halligan, P. (1987b). Development of a behavioral test of visuospatial neglect. Archives of Physical Medicine and Rehabilitation, 68(2), 98–102. Wurtz, R. H., Yamasaki, D. S., Duffy, C. J., & Roy, J. P. (1990). Functional specialization for visual motion processing in primate cerebral cortex. Cold Spring Harbor Symposia on Quantitative Biology, 55, 717–727. Zorzi, M., Mapelli, D., Rusconi, E., & Umilta, C. (2003). Automatic spatial coding of perceived gaze direction is revealed by the Simon effect. Psychonomic Bulletin and Review, 10(2), 423– 429.