Abstract

Despite an extensive literature on the neural substrates of semantic knowledge, how person-related information is represented in the brain has yet to be elucidated. Accordingly, in the present study we used functional magnetic resonance imaging (fMRI) to investigate the neural correlates of person knowledge. Focusing on the neural substrates of action knowledge, participants reported whether or not a common set of behaviors could be performed by people or dogs. While dogs and people are capable of performing many of the same actions (e.g. run, sit, bite), we surmised that the representation of this knowledge would be associated with distinct patterns of neural activity. Specifically, person judgments were expected to activate cortical areas associated with theory of mind (ToM) reasoning. The results supported this prediction. Whereas action-related judgments about dogs were associated with activity in various regions, including the occipital and parahippocampal gyri; identical judgments about people yielded activity in areas of prefrontal cortex, notably the right middle and medial frontal gyri. These findings suggest that person knowledge may be functionally dissociable from comparable information about other animals, with action-related judgments about people recruiting neural activity that is indicative of ToM reasoning.

Introduction

As other people are unquestionably the most important objects we encounter, it is surprising that little is currently known about the cortical representation of person-related knowledge. While neural regions that subserve important components of the person perception process (e.g. face perception, biological motion, emotion recognition, gaze detection) have been identified (Kanwisher et al., 1997; McCarthy et al., 1997; Puce et al., 1998; Allison et al., 2000; Hoffman and Haxby, 2000; Adolphs, 2003), research has yet to consider how abstract knowledge about other people is represented in the brain (see Mitchell et al., 2002). This oversight is puzzling as much is known about the cortical representation of object-based knowledge. Both neuropsychological and functional neuroimaging investigations have confirmed that the extraction of knowledge about particular classes of inanimate stimuli (e.g. tools, musical instruments, houses) is associated with distinct regions of the brain (Martin et al., 1996, 2000; Caramazza and Shelton, 1998; Chao et al., 1999; Caramazza, 2000; Haxby et al., 2000, 2001; Martin, 2001). But what of person knowledge?

One intriguing possibility is that the mind may have evolved dedicated neural circuits to deal with knowledge that has biological relevance or significance to people, such as information pertaining to plants, tools, and conspecifics (see Caramazza and Shelton, 1998; Caramazza, 2000). As Caramazza (2000, p. 1043) has argued,

evolutionary pressures resulted in specialized neural mechanisms for perceptually and conceptually distinguishing specific kinds of objects, leading to the categorical organization of knowledge in the brain. The range of categories that can be subsumed under this principle is very narrow—it is restricted to categories of objects whose rapid and accurate classification would have led to survival and reproductive advantages.

The benefits of such a distributed system lie in the rapid and accurate manner in which conceptual knowledge can be accessed. Were distinct classes of knowledge to share a common neuroanatomical location, inter-item competition (e.g. interference) may compromise the selection process. Any impediment to this process could have debilitating consequences in task contexts that necessitate the rapid generation of behavioral responses. Person perception is such a domain. The application of person knowledge demands a level of flexibility that is uncommon for most other classes of stimuli. Given its importance, one might expect fundamental aspects of person knowledge to be functionally dissociable from comparable classes of information in the brain (Caramazza, 2000), a possibility we explored in the current investigation.

To elucidate the architecture of person knowledge, we examined the neural substrates of one important class of information — human action. Despite the demonstration that object-related actions (e.g. kettleboil) are associated with activation in discrete regions of temporal cortex (see Petersen et al., 1988; Wise et al., 1991; Raichle et al., 1994; Martin et al., 1995; Poline et al., 1996; Warburton et al., 1996; Klein et al., 1999; Martin, 2001), studies have yet to identify cortical areas that support the representation of comparable person-based knowledge. Interestingly, some clues to the identity of these areas may be found in work that has investigated the neural correlates of mental state attribution (Baron-Cohen, 1995; Frith and Frith, 1999). Through the possession of a theory of mind (ToM), people are able to understand the goals and intentions that guide the behavior of others (Frith and Frith, 1999; Blakemore and Decety, 2001; Siegal and Varley, 2002; Gallagher and Frith, 2003). That is, people are able to infer the hidden internal states that give rise to human action.

Although distributed widely throughout the brain, the cerebral implementation of this capacity is known to involve a raft of frontal operations (Shallice, 2001). For example, observing action, adopting the perspective of another person, or inferring the mental state of an actor performing a social behavior produce patterns of activation in discrete regions of frontal cortex, notably the medial and middle frontal gyri (Fletcher et al., 1995a; Goel et al., 1995; Brunet et al., 2000; Castelli et al., 2000; Gallagher et al., 2000; Vogeley et al., 2001). These regions and their associated cognitive operations are believed to play a pivotal role in mindreading, the process through which people infer the motives underlying human action (Gallagher and Frith, 2003). By extension, might these frontal areas also play a prominent role in the representation of action-based knowledge (Frith and Frith, 1999)? That is, might the neural areas that support an agentic interpretation of human action also contribute to the representation of this knowledge in the brain?

To investigate this possibility, we used event-related functional magnetic resonance imaging (fMRI) to measure brain activity while participants made a series of action-related judgments about people and dogs. Dogs were selected as the target of comparison as a large number of actions can be performed by both dogs and people (e.g. run, jump, lick). We surmised, however, that these common behaviors would be associated with distinct patterns of neural activity when paired with each target. Specifically, the frontal activity that is indicative of ToM reasoning — notably, activity in medial prefrontal cortex (MPFC) — was only expected to emerge when the actions were paired with people. When the same actions were paired with dogs, activity was expected to emerge in occipitotemporal regions that support visual and memorial processing (i.e. bringing images of dogs to mind, see Cabeza and Nyberg, 2000; Kosslyn et al., 2001)

Materials and Methods

Participants

Twenty participants (12 female and 8 male, age range 18–33 years, mean age 23.8 years) completed the experiment. All participants were right-handed, native English speakers with no history of neurological problems. All gave informed consent according to the procedures approved by the Committee for the Protection of Human Subjects at Dartmouth College. Data from three participants (1 male and 2 females) were discarded due to problems with the acquisition of images during the functional runs.

Stimulus Materials and Procedure

The stimulus materials were 144 action words. Of these items, 72 comprised actions that could be performed by both people and dogs (e.g. leap, sit, swallow) and 72 comprised actions that could be performed by neither target (e.g. ferment, dissolve, ripple). Participants responded to visually presented words by pressing one of two response buttons (yes/no) ‘if the action could be performed by a person (or a dog) or could not be performed by a person (or a dog)’. On the person trials, the referent for the judgment task was a member of the opposite sex. Thus, female participants reported if a man could perform the actions and male participants reported if the actions could be performed by a woman. This instruction was included to prevent participants from using ‘self’ as the referent on the person trials. Debriefing confirmed that all participants followed the experimental instruction. Participants completed four runs (i.e. two person runs and two dog runs) and were instructed prior to each run whether the target was a person or a dog. Order of presentation of the runs was counterbalanced and the order of presentation of action words within each run was randomized for each participant. The words were presented using PsyScope (version 1.2.5) and back-projected with an Epson (ELP-7000) LCD projector onto a screen at the end of the magnet bore that participants viewed by way of a mirror mounted on the head coil. A pillow and foam cushions were placed within the head coil to minimize head movements. Each trial began with a fixation cross presented for 1000 ms. An action word (36 point Geneva font) was then presented for 1500 ms. Each fMRI run consisted of 36 trials in which the action could be performed by the target (i.e. person or dog) and 36 trials in which the action could not be performed by the target. A variable interstimulus interval (1000–8500 ms) was used to optimize estimation of the event-related BOLD response.

All images were collected using a 1.5 T GE Signa scanner with standard head coil. T1- weighted anatomical images were collected using a 3D sequence (SPGR; 128 sagittal slices, TR = 7 ms, TE = 3 ms, prep time = 315 ms, flip angle = 15°, FOV = 24 cm, slice thickness = 1.2 mm, matrix = 256 × 192). Functional images were collected with a gradient echo EPI sequence (each volume comprised 25 slices; 4.5 mm thick, 1 mm skip; TR = 2500 ms, TE = 35 ms, FOV = 24 cm, 64 × 64 matrix; 90° flip angle). Preprocessing and statistical analysis of the fMRI data were performed using SPM99 software (Wellcome Department of Cognitive Neurology, London). Preprocessing included slice timing and motion correction, coregistration to each participant’s anatomical data, normalization to the ICBM 152 brain template (Montreal Neurological Institute), and spatial smoothing with an 8 mm (full-width-at-half-maximum) Gaussian kernel. The hemodynamic response to the onset of each event type of interest was modeled with a canonical hemodynamic response function (HRF). Contrasts (t-tests, P < 0.001) for each voxel were run on individual participants and a random-effects model was then applied to the contrasts of all individual participants to obtain group activations.

Results

Behavioral data collected during scanning showed no difference in either the accuracy [t(16) < 1, nsMs 91% vs. 91%] or time taken [t(16) < 1, nsMs: 899 ms vs. 896 ms] by participants to make action judgments about people or dogs. Thus, it is unlikely that variability in the cortical activity associated with these judgments reflects differences in the difficulty of the respective tasks.

To examine whether action judgments about people and dogs were associated with different patterns of neural activity, we compared the event-related BOLD response associated with person trials to that associated with dog trials. This comparison revealed distinct patterns of brain activity for each type of target (see Table 1). Person > Dog comparisons yielded significant activations in the right middle frontal gyrus, right medial frontal gyrus and right anterior cingulate gyrus. In contrast, Dog > Person comparisons yielded significant activations in the left and right parahippocampal gyrus, left superior occipital gyrus, right. precuneus, left uncus, left cerebellum and left pulvinar. Figure 1 shows the significant activations that were observed in the right medial and middle frontal gyri when person and dog judgments were directly compared. The hemodynamic time courses for each region are also shown. In both brain regions, the temporal profile of activity was comparable. While person and dog judgments showed significant deactivations below the resting baseline (i.e. fixating a crosshair), the effect was most pronounced for action-based judgments about dogs. Thus, the apparent positive activation of these regions in the direct comparison results from the difference between two decreases relative to baseline. Figure 1 also shows the significant activations that were observed in the right parahippocampal gyrus and left pulvinar in the direct comparison between person and dog judgments. In both of these regions person and dog judgments yielded activations above the resting baseline, but the effect was most pronounced for dogs.

Discussion

The current results reveal that distinct areas of the brain subserve the representation of comparable action-related knowledge about people and other animals — namely dogs. Whereas action-related knowledge about dogs was associated with increased activity in occipitotemporal regions that support visual and memorial processing (i.e. visual imagery), person-based knowledge prompted activity in areas of right prefrontal cortex, areas that have previously been implicated in ToM reasoning and mental state attribution (Fletcher et al., 1995a; Goel et al., 1995; Brunet et al., 2000; Castelli et al., 2000; Gallagher et al., 2000; Vogeley et al., 2001). An extensive literature has documented the role of the frontal lobes in ToM reasoning (Frith and Frith, 1999). Notwithstanding the wide range of experimental paradigms that have been used to investigate this topic (e.g. verbal vs. non-verbal), a consistent pattern of results has emerged. The act of mentalizing routinely elicits activity in MPFC (Blakemore and Decety, 2001), activity that has been associated with people’s attempt to interpret social information in an agentic (i.e. goal-directed) manner.

Comparable processes appear to have been operating in the current investigation. Although people and dogs routinely perform identical actions (e.g. sit, bite, sniff), they typically do so for quite different reasons. Whereas dogs respond primarily to environmental cues, people act in accordance with internal goals, desires and intentions. Thus, observing a dog dash along the sidewalk is unlikely to elicit complex attributional processing. Witnessing a person perform the same action, however, is likely to trigger operations that furnish consciousness with candidate explanations for the behavior in question (e.g., perhaps the person has the goal of catching the bus). A basic feature of social-cognitive functioning may be that people continually strive to determine the hidden causes of human action (Frith and Frith, 1999). What is noteworthy about the present findings is that merely considering whether or not a person could perform particular actions was sufficient to activate frontal regions that have repeatedly been implicated in ToM reasoning (Blakemore and Decety, 2001; Siegal and Varley, 2002; Gallagher and Frith, 2003). In part, this further demonstrates the inherently social nature of the human mind and the ease with which the machinery of mental state attribution can be activated. Cortical areas that contribute to an agentic interpretation of human behavior (i.e. MPFC) also appear to be implicated in the representation of action-based knowledge in the brain (Frith and Frith, 1999; Gallagher and Frith, 2003).

It is important to note that the response observed in the MPFC in the current investigation differs from that which is typically referred to as an activation by researchers (i.e. an increase in activation relative to baseline). Responses in MPFC are almost always observed as decreases in activation relative to a resting baseline. While effects of this kind are generally difficult to interpret, recent work has attempted to elucidate why certain neural structures consistently deactivate in the presence of an active processing task (Shulman et al., 1997; Binder et al., 1999; Gusnard et al., 2001; Raichle et al., 2001; Stark and Squire, 2001). One possibility, noted by Raichle and colleagues (Raichle et al., 2001), is that structures such as the MPFC play a prominent role in default cognitive operations (e.g. stimulus-independent thought), hence deactivate when an external task is introduced. What is apparent from the present investigation, and other studies of this kind, is that MPFC plays a prominent role in fundamental aspects of social-cognitive functioning, such as the simulation of other minds, the use and representation of social knowledge, moral reasoning and self-referential mental activity (Anderson et al., 1999; Castelli et al., 2000; Adolphs, 2001; Milne and Grafman, 2001; Kelley et al., 2002; Mitchell et al., 2002). While these studies have typically reported increased activity in MPFC, it is important to note that these increases have consistently been observed relative to some other active task (Brunet et al., 2000; Castelli et al., 2000; Gallagher et al., 2000), thereby obscuring whether any changes occur relative to a resting baseline (but see Gusnard et al., 2001; Kelley et al., 2002; Mitchell et al., 2002). As observed in the current investigation, if compared to a resting baseline task (e.g. fixating a crosshair), ToM reasoning may be associated with deactivations in MPFC.

In the only other study to date that has investigated the representation of abstract person-based knowledge, Mitchell et al. (2002) also reported activity in MPFC. These researchers speculated that activity in this region may subserve the specialized cognitive operations that are brought to bear when people think about conspecifics, such as when they represent the internal states of other social agents (Frith and Frith, 1999). The present findings are supportive of this viewpoint. The differential neural activity that was associated with action-related judgments about people and dogs may reflect the different cognitive strategies that participants use when they think about (i.e. judge) each type of target. Specifically, action judgments about dogs yielded activity in areas that have commonly been associated with memory-guided visual imagery (e.g. superior occipital gyrus, parahippocampal gyrus, precuneus — Fletcher et al., 1995b; Cabeza and Nyberg, 2000; Ishai et al., 2000, 2002; Jeannerod, 2001; Kosslyn et al., 2001), suggesting that task performance may have been supported by this mental operation. Person judgments, in contrast, did not rely on mental imagery to the same degree, instead they were associated with prefrontal operations that are indicative of ToM reasoning (Blakemore and Decety, 2001; Siegal and Varley, 2002). Thus, whereas participants visualized animal (i.e. dog) action, they appeared to reason about comparable human behavior. This does not imply, however, that judgments about animals would never elicit ToM reasoning. One interesting possibility is that animal actions could enlist ToM reasoning, at least for certain individuals. Anecdotal evidence suggests that dog owners routinely construe canine behavior in an intentional, goal-directed manner. Thus, when considering the actions of a pet, one might expect these individuals to show modulations of MPFC that are indicative of mental state attribution (i.e., the intentional stance).

One of the most oft-reported findings in cognitive neuroscience is that different classes of knowledge are associated with distinct regions of the brain (Martin et al., 2000; Martin, 2001). Although the exact nature of this neuroanatomical localization remains open to debate, most researchers concur that the brain contains some kind of category-specific neural architecture. Based on this observation, some theorists have suggested that the mind may have evolved dedicated neural circuits to deal with fundamental aspects of social-cognitive functioning, such as person perception and the representation of person-based knowledge (Baron-Cohen, 1995; Caramazza and Shelton, 1998; Caramazza, 2000;). Supporting this general viewpoint, recent research has shown that distinct neural systems subserve the representation of person and object knowledge — specifically, attributes that are descriptive of people and objects (Mitchell et al., 2002). As demonstrated in the present inquiry, the seemingly ‘special’ status of person knowledge may also extend to the representation of abstract action-related information. Compared to inferences about actions performed by dogs, identical judgments about human acts were accompanied by neural activity that is indicative of ToM reasoning. Construing human action in a mentalistic manner may therefore be an obligatory facet of social-cognitive functioning.

The authors thank Bill Kelley, Joe Moran, Gagandeep Wig, Abby Baird, Heather Gordon, Dave Turk and Arie van der Lugt for their help with this work.

Figure 1. Activation map showing brain areas to be more active during Person than Dog trials (red) and Dog than Person trials (blue) from the group analysis (n = 17). Regions of modulation included the (A) right medial frontal gyrus (10, 48, 34), (B) right middle frontal gyrus (34, 35, 35), (C) right parahippocampal gyrus (24, –46, 10) and (D) left pulvinar (–24, –29, 14). The hemodynamic time courses associated with these areas are also depicted. Time courses (expressed as percent signal change relative to baseline) were computed for each condition within a 3D region surrounding the peak voxel identified from the contrast statistical map (people vs. dogs). Regions were defined using an automated algorithm that identified all contiguous voxels within 6 mm of the peak that reached the significance level (P < 0.0001, uncorrected).

Figure 1. Activation map showing brain areas to be more active during Person than Dog trials (red) and Dog than Person trials (blue) from the group analysis (n = 17). Regions of modulation included the (A) right medial frontal gyrus (10, 48, 34), (B) right middle frontal gyrus (34, 35, 35), (C) right parahippocampal gyrus (24, –46, 10) and (D) left pulvinar (–24, –29, 14). The hemodynamic time courses associated with these areas are also depicted. Time courses (expressed as percent signal change relative to baseline) were computed for each condition within a 3D region surrounding the peak voxel identified from the contrast statistical map (people vs. dogs). Regions were defined using an automated algorithm that identified all contiguous voxels within 6 mm of the peak that reached the significance level (P < 0.0001, uncorrected).

Table 1


 Regions of increased brain activity associated with accessing action knowledge

    t-value  
Anatomical label x y z Person Dog 
Person > Dog activations      
Right middle frontal gyrus (BA 9)    34   35   35  –4.18* –5.52* 
Right middle frontal gyrus (BA 6/8)    26   16   45  –5.35* –6.92* 
Right medial frontal gyrus (BA 8/9)   10   48   34  –5.15* –8.18* 
Right anterior cingulate gyrus (BA 32)    6   20   40    5.06*  2.71 
      
Dog > Person activations      
Right parahippocampal gyrus (BA 19/30)  24 –46   10    2.58  6.43* 
Left parahippocampal gyrus (BA 19) –28 –52    4    3.18  6.73*  
Left pulvinar  –24 –29   14    1.88  4.63* 
Left pulvinar –14 –32   16    3.45  5.88* 
Left superior occipital gyrus (BA 19) –40 –78    33 –11.39* –7.33* 
Right precuneus (BA 31)  26 –71   18  –5.87* –5.22* 
Left cerbellum –12 –39  –8    0.46  5.07* 
Left uncus (BA 28) –12    6 –26    1.31  6.32* 
    t-value  
Anatomical label x y z Person Dog 
Person > Dog activations      
Right middle frontal gyrus (BA 9)    34   35   35  –4.18* –5.52* 
Right middle frontal gyrus (BA 6/8)    26   16   45  –5.35* –6.92* 
Right medial frontal gyrus (BA 8/9)   10   48   34  –5.15* –8.18* 
Right anterior cingulate gyrus (BA 32)    6   20   40    5.06*  2.71 
      
Dog > Person activations      
Right parahippocampal gyrus (BA 19/30)  24 –46   10    2.58  6.43* 
Left parahippocampal gyrus (BA 19) –28 –52    4    3.18  6.73*  
Left pulvinar  –24 –29   14    1.88  4.63* 
Left pulvinar –14 –32   16    3.45  5.88* 
Left superior occipital gyrus (BA 19) –40 –78    33 –11.39* –7.33* 
Right precuneus (BA 31)  26 –71   18  –5.87* –5.22* 
Left cerbellum –12 –39  –8    0.46  5.07* 
Left uncus (BA 28) –12    6 –26    1.31  6.32* 

Coordinates are from the Talairach and Tournoux (1998) atlas. Person and Dog columns display the t-value associated with the area’s peak hemodynamic response relative to passive baseline for Person and Dog trials, respectively: *P < 0.001.

References

Adolphs R (
2001
) The neurobiology of social cognition.
Curr Opin Neurobiol
 
11
:
231
–239.
Adolphs R (
2003
) Cognitive neuroscience of human social behaviour.
Nat Rev Neurosci
 
4
:
165
–178.
Allison T, Puce A, McCarthy G (
2000
) Social perception from visual cues: role of the STS region.
Trends Cogn Sci
 
4
:
267
–278.
Anderson SW, Bechara A, Damasio H, Tranel D, Damasio AR (
1999
) Impairment of social and moral behavior related to early damage in human prefrontal cortex.
Nat Neurosci
 
2
:
1032
–1037.
Baron-Cohen S (
1995
) Mindblindness: an essay on autism and theory of mind. Cambridge, MA: MIT Press.
Binder JR, Frost JA, Hammeke TA, Bellgowan PSF, Rao SM, Cox RW (
1999
) Conceptual processing during the conscious resting state: a functional MRI study.
J Cogn Neurosci
 
11
:
80
–93.
Blakemore, SJ and Decety, J (
2001
) From the perception of action to the understanding of intention.
Nat Rev Neurosci
 
2
:
561
–567.
Brunet E, Sarfati Y, Hardy-Bayle MC, Decety J (
2000
) A PET investigation of the attribution of intentions with a nonverbal task.
Neuroimage
 
11
:
157
–66.
Cabeza, R and Nyberg, L (
2000
) Imaging cognition II: An empirical review of 275 PET and fMRI studies.
J Cogn Neurosci
 
12
:1-47.
Caramazza A (
2000
) The organization of conceptual knowledge in the brain. In: The new cognitive neurosciences (Gazzaniga MS, ed.), pp.
1037
–1046. Cambridge, MA: MIT Press.
Caramazza, A and Shelton, JR (
1998
) Domain-specific knowledge systems in the brain: the animate–inanimate distinction.
J Cogn Neurosci
 
10
:
1
–34.
Castelli F, Happé F, Frith U, Frith C (
2000
) Movement and mind: a functional imaging study of perception and interpretation of complex intentional movement patterns.
Neuroimage
 
12
:
314
–325.
Chao LL, Haxby JV, Martin A (
1999
) Attribute-based neural substrates in temporal cortex for perceiving and knowing about objects.
Nat Neurosci
 
2
:
913
–919.
Fletcher PC, Happé F, Frith U, Baker SC, Dolan RJ, Frackowiak RSJ, Frith CD (
1995
) Other minds in the brain: a functional imaging study of ‘theory of mind’ in story comprehension.
Cognition
 
57
:
109
–128.
Fletcher PC, Frith CD, Baker SC, Shallice T, Frackowiak RSJ, Dolan RJ (
1995
) The mind’s eye — precuneus activation in memory-related imagery.
Neuroimage
 
2
:
195
–200.
Frith, CD and Frith, U (
1999
) Interacting minds — a biological basis.
Science
 
286
:
1692
–1695.
Gallagher, HL and Frith, CD (
2003
) Functional imaging of ‘theory of mind’.
Trends Cogn Sci
 
7
:
77
–83.
Gallagher HL, Happé F, Brunswick N, Fletcher PC, Frith U, Frith CD (
2000
) Reading the mind in cartoons and stories: an fMRI study of ‘theory of mind’ in verbal and nonverbal tasks.
Neuropsychologia
 
38
:
11
–21.
Goel V, Grafman J, Sadato N, Hallett M (
1995
) Modeling other minds.
Neuroreport
 
6
:
1741
–1746.
Gusnard DA, Akbudak E, Shulman GL, Raichle ME (
2001
) Medial prefrontal cortex and self-referential mental activity: relation to a default mode of brain function.
Proc Natl Acad Sci USA
 
98
:
4259
–4264.
Haxby JV, Ishai A, Chao L, Ungerleider LG, Martin A (
2000
) Object form topology in the ventral temporal lobe.
Trends Cogn Sci
 
4
:
3
–4.
Haxby JV, Gobbini MI, Furey ML, Ishai A, Schouten JL, Pietrini P (
2001
) Distributed and overlapping representations of faces and objects in ventral temporal cortex.
Science
 
293
:
2425
–2430.
Hoffman, EA and Haxby, JV (
2000
) Distinct representations of eye gaze and identity in the distributed human neural system for face processing.
Nat Neurosci
 
3
:
80
–84.
Ishai A, Underleider LG, Haxby JV (
2000
) Distributed neural systems for the generation of visual images.
Neuron
 
28
:
979
–990.
Ishai A, Haxby JV, Ungerleider LG (
2002
) Visual imagery of famous faces: effects of memory and attention revealed by fMRI.
Neuroimage
 
17
:
1729
–1741.
Jeannerod M (
2001
). Neural simulation of action: a unifying mechanism for motor cognition.
Neuroimage
 
14
:
103
–109.
Kanwisher N, McDermott J, Chun MM (
1997
) The fusiform face area: a module in human extrastriate cortex specialized for face perception.
J Neurosci
 
17
:
4302
–4311.
Kelley WM, Macrae CN, Wyland CL, Caglar S, Inati S, Heatherton TF (
2002
) Finding the self? An event-related fMRI study.
J Cogn Neurosci
 
14
:
785
–794.
Klein D, Olivier A, Milner B, Zatorre RJ, Zhao V, Nikelski J (
1999
) Cerebral organization in bilinguals: a PET study of Chinese-English verb generation.
Neuroreport
 
10
:
2841
–2846.
Kosslyn SM, Ganis G, Thompson WL (
2001
) Neural foundations of imagery.
Nat Rev Neurosci
 
2
:
635
–642.
Martin A (
2001
) Functional neuroimaging of semantic memory. In: Handbook of functional neuroimaging of cognition (Cabeza R, Kingstone A, eds), pp.
153
–186. Cambridge, MA: MIT Press.
Martin A, Haxby JV, Lalonde FM, Wiggs CL, Ungerleider LG (
1995
) Discrete cortical regions associated with knowledge of color and knowledge of action.
Science
 
270
:
102
–105.
Martin A, Wiggs CL, Ungerleider LG, Haxby JV (
1996
) Neural correlates of category-specific knowledge.
Nature
 
379
:
649
–652.
Martin A, Ungerleider LG, Haxby JV (
2000
) Category-specificity and the brain: the sensory-motor model of semantic representations of objects. In: The new cognitive neurosciences (Gazzaniga MS, ed.), pp.
1023
–1036. Cambridge, MA: MIT Press.
McCarthy G, Puce A, Gore JC, Allison T (
1997
) Face-specific processing in the fusiform gyrus.
J Cogn Neurosci
 
9
:
605
–610.
Milne, E and Grafman, J (
2001
) Ventromedial prefrontal cortex lesions in humans eliminate implicit gender stereotyping.
J Neurosci
 
21
:RC150.
Mitchell JP, Heatherton TF, Macrae CN (
2002
) Distinct neural systems subserve person and object knowledge.
Proc Natl Acad Sci USA
 
99
:
15238
–15243.
Petersen SE, Fox PT, Posner MI, Mintun MA, Raichle ME (
1988
) Positron emission tomographic studies of the cortical anatomy of single-word processing.
Nature
 
331
:
585
–589.
Poline JB, Vandenberghe R, Holmes AP, Friston KJ, Frackowiak RS (
1996
) Reproducibility of PET activation studies: lessons from a multi-center European experiment.
Neuroimage
 
4
:
34
–54.
Puce A, Allison T, Bentin S, Gore JC, McCarthy G (
1998
) Temporal cortex activation in humans viewing eye and mouth movements.
J Neurosci
 
18
:
2188
–2199.
Raichle ME, Fiez JA, Videen TO, MacLeod AM, Pardo JV, Fox PT, Petersen SE (
1994
). Practice-related changes in human brain functional anatomy during nonmotor learning.
Cereb Cortex
 
4
:
8
–26.
Raichle ME, MacLeod AM, Snyder AZ, Powers WJ, Gusnard DA, Shulman GL (
2001
) A default mode of brain function.
Proc Natl Acad Sci USA
 
98
:
676
–682.
Shallice T (
2001
) ‘Theory of mind’ and the prefrontal cortex.
Brain
 
124
:
247
–248.
Shulman GL, Fiez JA, Corbetta M, Buckner RL, Miezin FM, Raichle ME, Petersen SE (
1997
) Common blood flow changes across visual tasks: II., Decreases in cerebral cortex.
J Cogn Neurosci
 
9
:
648
–663.
Siegal, M and Varley, R (
2002
) Neural systems involved in ‘theory of mind’.
Nat Rev Neurosci
 
3
:
463
–471.
Stark, CE and Squire, LR (
2001
) When zero is not zero: the problem of ambiguous baseline conditions in fMRI.,
Proc Natl Acad Sci USA
 
98
:
12760
–12766.
Talairach, J and Tournoux, P (
1998
) Co-planar stereotaxic atlas of the human brain. New York: Thieme Verlag.
Vogeley K, Bussfield P, Newen A, Herrmann S, Happe F, Falkai P, Maier W, Shah NJ, Fink GR, Zills K (
2001
) Mind reading: neural mechanisms of theory of mind and self-perspective.
Neuroimage
 
14
:
170
–181.
Warburton E, Wise RJ, Price CJ, Weiller C, Hadar U, Ramsay S, Frackowiak RS (
1996
) Noun and verb retrieval by normal subjects. Studies with PET.,
Brain
 
119
:
159
–179.
Wise R, Chollet F, Hadar U, Friston K, Hoffner E, Frackowiak R (
1991
) Distribution of cortical neural networks involved in word comprehension and word retrieval.
Brain
 
114
:
1803
–1817.