{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,17]],"date-time":"2026-10-17T08:13:42Z","timestamp":1792224822106,"version":"4.4.0"},"reference-count":35,"publisher":"American Association for the Advancement of Science (AAAS)","issue":"5539","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Science"],"published-print":{"date-parts":[[2001,9,28]]},"abstract":"<jats:p>Despite extensive evidence for regions of human visual cortex that respond selectively to faces, few studies have considered the cortical representation of the appearance of the rest of the human body. We present a series of functional magnetic resonance imaging (fMRI) studies revealing substantial evidence for a distinct cortical region in humans that responds selectively to images of the human body, as compared with a wide range of control stimuli. This region was found in the lateral occipitotemporal cortex in all subjects tested and apparently reflects a specialized neural system for the visual perception of the human body.<\/jats:p>","DOI":"10.1126\/science.1063414","type":"journal-article","created":{"date-parts":[[2002,7,27]],"date-time":"2002-07-27T05:39:52Z","timestamp":1027748392000},"page":"2470-2473","source":"Crossref","is-referenced-by-count":1790,"title":["A Cortical Area Selective for Visual Processing of the Human Body"],"prefix":"10.1126","volume":"293","author":[{"given":"Paul E.","family":"Downing","sequence":"first","affiliation":[{"name":"School of Psychology, Centre for Cognitive Neuroscience, University of Wales, Bangor LL57 2AS, UK."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuhong","family":"Jiang","sequence":"additional","affiliation":[{"name":"Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139, USA."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Miles","family":"Shuman","sequence":"additional","affiliation":[{"name":"Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139, USA."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nancy","family":"Kanwisher","sequence":"additional","affiliation":[{"name":"Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139, USA."},{"name":"Massachusetts General Hospital\u2013Nuclear Magnetic Resonance Imaging Center, 149 13th Street, Charlestown, MA 02129, USA."}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"221","reference":[{"key":"e_1_3_1_2_2","doi-asserted-by":"publisher","DOI":"10.1038\/77664"},{"key":"e_1_3_1_3_2","doi-asserted-by":"publisher","DOI":"10.1038\/72140"},{"key":"e_1_3_1_4_2","doi-asserted-by":"publisher","DOI":"10.1080\/13554799808410631"},{"key":"e_1_3_1_5_2","doi-asserted-by":"publisher","DOI":"10.1162\/089892900562417"},{"key":"e_1_3_1_6_2","doi-asserted-by":"publisher","DOI":"10.1523\/JNEUROSCI.18-06-02188.1998"},{"key":"e_1_3_1_7_2","doi-asserted-by":"publisher","DOI":"10.1016\/S0960-9822(02)00646-2"},{"key":"e_1_3_1_8_2","doi-asserted-by":"crossref","unstructured":"G. Le Clec'H et al. Neuroimage 12 381 (2000).","DOI":"10.1006\/nimg.2000.0627"},{"key":"e_1_3_1_9_2","doi-asserted-by":"publisher","DOI":"10.1093\/cercor\/4.5.509"},{"key":"e_1_3_1_10_2","doi-asserted-by":"publisher","DOI":"10.1006\/brcg.2000.1231"},{"key":"e_1_3_1_11_2","unstructured":"Scanning was conducted on a 1.5 T (experiments 1 to 6) or 3.0 T (experiment 7) Siemens scanner at the Massachusetts General Hospital\u2013NMR center in Charlestown MA. The general procedure was identical in all experiments. A gradient echo pulse sequence (TR = 2 s) was used. Sixteen to 20 slices (4 to 6 mm thick 3.125 mm by 3.125 mm in-plane resolution) were collected using a standard head coil. The slices were either parallel to the brainstem starting near the occipital pole and extending anteriorly to cover most of the occipital and parietal lobes and at least the posterior half of the temporal lobe or axial covering most of the cortex. Six subjects participated in experiment 1 and experiment 2 in the same session; of these four also participated in experiment 3 in the same session. Five new subjects participated in experiments 4 and 5 in the same session. Six new subjects participated in experiment 6 (one of whom had participated in earlier studies) and three new subjects participated in experiment 7. In each experiment photographs of human body parts and of object parts were included in separate blocks as baselines and as localizers for other scans. The remaining conditions in each experiment were as follows. Experiment 1: faces face parts hands (in a single configuration) whole objects scrambled objects and scenes; experiment 2: faces hands (in multiple configurations) oscillating low-contrast rings static low-contrast rings line drawings of body parts and line drawings of object parts; experiment 3: human bodies and mammals both with heads removed in separate blocks; experiment 4: photographs and line drawings of humans both alert and inactive and line drawings and photographs of cars; experiment 5: humans with heads removed mammals birds fish trees and articulated objects; experiment 6: human silhouettes and scrambled versions human stick figures and scrambled versions empty and filled clothes; experiment 7: humans mammals chairs tools faces and scenes. The stimuli in all experiments were gray-scale photographs (or line drawings where applicable) on a uniform background. In each experiment 15 exemplars from each category were tested. Each scan consisted of 16 experimental blocks (16 s each) with a 16-s fixation-only period intervening every four blocks and at the beginning and end of the scan. Twenty images were presented individually within each block each appearing for 300 ms followed by a fixation-only interval of 500 ms. Block order was counterbalanced between two versions of each experiment so that mean serial position in the scan was equivalent across conditions. Subjects were typically scanned four times on each experiment; scans from different experiments were interleaved in the same session. Subjects performed a 1-back repetition detection task during the scans pressing a button whenever the same stimulus appeared twice in immediate succession which occurred twice per block."},{"key":"e_1_3_1_12_2","unstructured":"Regions of interest (ROIs) were defined as contiguous voxels in the occipitotemporal area of the right hemisphere where the response to human bodies\/body parts was significantly greater than to controls at P < 10 \u22124 (uncorrected) on a Kolmogorov-Smirnov test. Many subjects also showed a substantially weaker but significant activation in the left hemisphere. The response properties of this left hemisphere ROI were measured where possible and found to closely match the right hemisphere; only the right hemisphere data are reported here. For analyses concerning overlap with other cortical areas (described below) both hemispheres were considered. We also determined the average (with standard deviation) across seven subjects of the MNI coordinates of the most significant voxel in the EBA: [51 (2) \u201371 (4) 1 (3)] for the right hemisphere and [\u201351 (4) \u201372 (5) 8 (4)] for the left hemisphere. [We report MNI coordinates (based on the Montreal Neurological Institute (MNI) probabilistic brain atlas) as they are based on the average of many brains rather than a single one and they are the native coordinate system of the widely used SPM package.]"},{"key":"e_1_3_1_13_2","unstructured":"We focus here on the visual properties of this extrastriate region. These results do not exclude the possibility that stimulation in other modalities could activate the EBA (30). Other questions for future research include the possible role of the EBA in representing general amodal semantic knowledge about the body and the relation between the EBA and parietal and frontal systems involved in the representation and generation of actions (31)."},{"key":"e_1_3_1_14_2","unstructured":"Mean time courses were extracted for each subject in each experiment averaging across scans and across the voxels in the ROI. Percent signal change was determined for each subject and each condition relative to the fixation-only baseline epochs. Hemodynamic lag and mixing of effects at epoch boundaries were accounted for by assigning the first time point of each epoch to the previous condition and dropping the subsequent time point. All reported data reflect grand mean percent signal change from fixation across subjects. The data used to calculate the percent signal changes in each condition were collected from different scans than those used to identify the ROIs."},{"key":"e_1_3_1_15_2","unstructured":"The low response to faces distinguishes the EBA from other apparently face-selective regions of the STS (32)."},{"key":"e_1_3_1_16_2","unstructured":"It is not clear whether the higher response to whole bodies reflects a preference for whole bodies over body parts or the presence in the stimulus of multiple undetached body parts. Both possibilities however are consistent with a representation in the EBA of the appearance of the human body."},{"key":"e_1_3_1_17_2","unstructured":"In experiment 3 we tested human bodies with heads removed to account for the possibility suggested by the low response to faces alone that faces might inhibit activation of the EBA. Whereas in experiment 3 the response to (headless) bodies was greater than to body parts in experiment 4 the response to bodies with intact faces (1.5%) did not differ from that to body parts (1.5%) consistent with this inhibition hypothesis."},{"key":"e_1_3_1_18_2","unstructured":"Although a significant interaction between object type (body versus nonbody) and format (photograph versus line drawing) was found in experiment 2 ( P < 0.05) this interaction did not reach significance in experiment 4 ( P > 0.1)."},{"key":"e_1_3_1_19_2","unstructured":"As bodies (unlike many man-made objects) have the capacity to move on their own we also tested whether the implication of motion is necessary to produce the high EBA response to bodies observed here. In experiment 4 we compared photographs and line drawings of alert upright humans with control pictures of the same individuals lying down with their eyes closed as if asleep (33). No significant difference was found in the EBA response to alert (1.6%) compared with inactive (1.6%) people. Thus the implication of motion is not necessary to produce a strong response in the EBA."},{"key":"e_1_3_1_20_2","doi-asserted-by":"publisher","DOI":"10.1038\/13217"},{"key":"e_1_3_1_21_2","unstructured":"The FFA was identified with a contrast comparing faces with objects; the PPA by comparing scenes with faces and objects; LOC by comparing intact with scrambled objects; and MT\/V5 by comparing moving with static low-contrast rings. The FFA and PPA were localized with the relevant contrasts from experiment 1. The LOC MT and retinotopic areas were localized in separate scans concurrent with experiments 4 5 and 7. Five additional subjects were scanned on LOC MT and EBA localizers to contribute to this analysis."},{"key":"e_1_3_1_22_2","unstructured":"Retinotopic regions were identified with contrast-reversing patterned rings appearing at three different eccentricities surrounding fixation."},{"key":"e_1_3_1_23_2","doi-asserted-by":"publisher","DOI":"10.1523\/JNEUROSCI.17-11-04302.1997"},{"key":"e_1_3_1_24_2","doi-asserted-by":"publisher","DOI":"10.1038\/33402"},{"key":"e_1_3_1_25_2","doi-asserted-by":"publisher","DOI":"10.1523\/JNEUROSCI.20-09-03310.2000"},{"key":"e_1_3_1_26_2","doi-asserted-by":"publisher","DOI":"10.1523\/JNEUROSCI.15-04-03215.1995"},{"key":"e_1_3_1_27_2","unstructured":"On average 31% of EBA voxels (median 21%) were also significantly activated in either the MT or LOC localizers; in no subject was the overlap complete. There are several reasons why objectively discrete neural systems might appear to overlap as measured with fMRI. First functionally independent neural networks might occupy the same region of cortex in an interdigitated fashion. Second the boundary between two networks that are spatially discrete but adjacent is unlikely to fall exactly along voxel borders. Third limits to the spatial resolution of the BOLD signal reduce the independence of the signal from adjacent voxels. As a result fMRI methods cannot support definitive claims about the absence of overlap between adjacent neural systems. Thus the possibility of partial overlap between the EBA and MT\/LOC cannot be decisively ruled out. However the present evidence demonstrates that the EBA cannot be a complete subset of the other visual regions tested."},{"key":"e_1_3_1_28_2","unstructured":"We also considered the response of the EBA to biological motion displays (34). The independently localized EBA responded significantly more to point-light animations of biological motion than to scrambled controls ( P < 0.025; n = 10). This result is consistent with our other findings as the experimental stimulus suggests the presence of a human figure whereas the control does not. We further tested the response of the EBA ( n = 5) to gray-scale images of human bodies and articulated objects presented in short animation sequences. The motion sequences were either \u201cartificial \u201d in which the entire object appeared to oscillate rigidly around its midpoint or \u201cnatural \u201d in which the object moved in a realistic manner (e.g. a kicking motion in the human-object condition or scissors opening and closing in the articulated-object condition). The EBA response was greater to natural motion than to artificial motion for both human bodies (2.2% versus 2.0%; P < 0.025) and for objects (1.2% versus 0.8%; P < 0.01). There were significant main effects of both motion type ( P < 0.005) and object type ( P < 0.001) but critically these two factors did not reliably interact. If anything the increase in response to natural motion over artificial motion was greater for objects (0.4%) than for humans (0.3%). Further the higher response for natural motion compared with artificial motion was also observed in area MT (in voxels that did not overlap with the EBA); for bodies (2.7% versus 2.3%; P < 0.05) and objects (2.4% versus 1.8%; P < 0.001) and this effect did not significantly differ from the same effect seen in the EBA. These findings in both the EBA and MT do not depend on using artificial motion as the control; the same pattern of data and statistical signficance held when natural motion was compared with static images. This suggests a generalized (perhaps attentional) enhancement of the response to the natural-motion stimuli whether bodies or objects. Thus although the EBA is sensitive to motion information it has no specific preference for biological motion per se."},{"key":"e_1_3_1_29_2","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.92.18.8135"},{"key":"e_1_3_1_30_2","doi-asserted-by":"publisher","DOI":"10.1016\/S0042-6989(01)00073-6"},{"key":"e_1_3_1_31_2","doi-asserted-by":"publisher","DOI":"10.1038\/85201"},{"key":"e_1_3_1_32_2","doi-asserted-by":"publisher","DOI":"10.1016\/S0959-4388(97)80037-2"},{"key":"e_1_3_1_33_2","doi-asserted-by":"publisher","DOI":"10.1523\/JNEUROSCI.16-16-05205.1996"},{"key":"e_1_3_1_34_2","doi-asserted-by":"publisher","DOI":"10.1162\/08989290051137594"},{"key":"e_1_3_1_35_2","doi-asserted-by":"publisher","DOI":"10.3758\/BF03212378"},{"key":"e_1_3_1_36_2","unstructured":"We thank R. Epstein K. Grill-Spector M. Hoerger Z. Kourtzi R. Saxe D. Stanley and the rest of the Kanwisher lab for helpful discussions and suggestions L. Spelke A. Wagner and S. Pinker for comments on the manuscript and E. Grossman and R. Blake for providing the biological motion stimuli. Supported by a Human Frontiers grant to N.K."}],"container-title":["Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.science.org\/doi\/pdf\/10.1126\/science.1063414","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,1,9]],"date-time":"2024-01-09T17:22:06Z","timestamp":1704820926000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.science.org\/doi\/10.1126\/science.1063414"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2001,9,28]]},"references-count":35,"journal-issue":{"issue":"5539","published-print":{"date-parts":[[2001,9,28]]}},"alternative-id":["10.1126\/science.1063414"],"URL":"https:\/\/doi.org\/10.1126\/science.1063414","relation":{"has-review":[{"id-type":"doi","id":"10.3410\/f.1001956.11859","asserted-by":"object"}]},"ISSN":["0036-8075","1095-9203"],"issn-type":[{"value":"0036-8075","type":"print"},{"value":"1095-9203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2001,9,28]]}}}