PSYCHOLOGICAL SCIENCE

Research Article THE PERCEIVED ONSET TIME OF SELF- AND OTHER-GENERATED ACTIONS Andreas Wohlschläger,1 Patrick Haggard,2 Benno Gesierich,3 and Wolfgang Prinz1 1

Max-Planck-Institut für Psychologische Forschung, München, Germany; 2University College London, London, United Kingdom; and 3Università degli Studi di Ferrara, Ferrara, Italy

Abstract—Awareness of actions is partly based on the intentions accompanying them. Thus, the awareness of self- and other-generated actions should differ to the extent that access to own and other’s intentions differs. Recent studies have found a brain circuit (the mirrorneuron system) that represents self- and other-generated actions in an integrated fashion. This system does not respond to actions made by nonagents, such as machines. We measured the estimated onset time of actions that subjects either executed themselves or observed being executed by someone else or by a machine. In three experiments, the estimates of the machine actions always differed from those of self- and other-generated actions, whereas the latter two were indistinguishable. Our results are consistent with the view that intentions are attributed to others but not to machines. They also raise the interesting possibility that people attribute intentions to themselves in the same way as they do to others. Voluntary action is fundamental to human existence. The traditional concept of free will views action as starting within the mind of the individual. Neuroscientific studies have identified a specific role of frontal motor areas in the generation of voluntary actions. Kornhuber and Deecke (1965) identified a Bereitschaftspotential (readiness potential) over frontal brain regions, beginning at least 1 s before a voluntary action, and growing gradually to a clear maximum just before movement. Later studies identified the supplementary motor area as a key structure in the internal generation of actions (Goldberg, Kwan, Borrett, & Murphy, 1984; Tanji, 2001) and a likely source of the readiness potential (Kornhuber & Deecke, 1965). These studies all point to the frontal motor areas as a key circuit for willed actions. Additional studies have investigated the relations between neural preparation of action and conscious awareness (“free will”). Libet, Gleason, Wright, and Pearl (1983) found that the readiness potential preceding voluntary action began 300 to 500 ms before subjects became aware that they intended to move, in apparent contradiction to the traditional Cartesian concept of free will. The connection between frontal brain activity and conscious experience of action was confirmed by Fried et al. (1991). When they stimulated frontal cortex intracranially at low intensity, their patients reported an urge to move a specific body part. Higher intensities of stimulation evoked actual movements of the same body part. These studies present voluntary actions as a specific brain function linked to subjective mental life. We call this the privacy position. In contrast to this subjective tradition, there is increasing evidence that a common brain circuit is used both to control object-oriented actions and to represent homologous actions of others. Thus, mirror neurons in the monkey premotor cortex fire both during grasping and

Address correspondence to Andreas Wohlschläger, Max Planck Institute for Psychological Research, Cognition and Action, Amalienstraße 33, D-80799 München, Germany; e-mail: [email protected].

586

Copyright © 2003 American Psychological Society

when the monkey observes a human or conspecific make a similar grip (Gallese, Fadiga, Fogassi, & Rizzolatti, 1996). However, these neurons are silent when kinematically similar actions are performed using a tool. Functional imaging also confirms activation of a parietal premotor network in both the control and the observation of action in humans (Rizzolatti, Fadiga, Matelli, et al., 1996). This evidence suggests that the neural representation of voluntary action is not private and subjective, but integrates across agents, and is thus intrinsically social (Gallese, 2000). Moreover, observing the actions of others can directly influence the action system of the observer. Fadiga, Fogassi, Pavesi, and Rizzolatti (1995) found increased excitability of the motor cortex to transcranial magnetic stimulation when subjects observed an action (see also Aziz-Zadeh, Maeda, Zaidel, Mazziota, & Iacoboni, 2002; Maeda, Kleiner-Fisman, & Pascual-Leone, 2002). We use the term integrationist to refer to this tradition in which action representation is seen as comparable across agents. Recent evidence from both human performance (Craighero, Fadiga, Rizzolatti, & Umiltà, 1999; Craighero, Fadiga, Umiltà, & Rizzolatti, 1996) and neurophysiological (Gallese et al., 1996; Rizzolatti, Fadiga, Gallese, & Fogassi, 1996) studies supports an integrationist position. But, to our knowledge, no psychophysical work has compared awareness of self-generated actions with awareness of other-generated actions. Privacy and integrationist accounts of action make different predictions about these situations. According to the privacy account, one’s awareness of other people’s actions must be different from one’s awareness of one’s own actions, because one has special access to only one’s own intentions. According to the integrationist account, one’s awareness of other people’s actions should be fundamentally similar to one’s awareness of one’s own actions. A radical integrationism could reject privacy entirely by claiming that one has no better access to one’s own intentions than to the intentions of others. For example, one might retrospectively infer intentions from actions in both these cases. In summary, then, the private view contrasts self-initiated actions with all other events. The integrationist view contrasts actions (irrespective of the agent) with events not involving an agent. We performed a series of three experiments aimed at exploring the boundaries between You, Me, and It. We used the method devised by Libet et al. (1983) because the resulting measure of the perceived onset time of actions provides a common metric for describing diverse events, yet is also sensitive to the properties of the underlying psychological representations of those events (Haggard, Aschersleben, Gehrke, & Prinz, 2002).

EXPERIMENT 1 In Experiment 1, we compared the perceived onset time of voluntary actions made by the subject with the perceived onset time of actions that the subject observed the experimenter make and of comparable mechanical events that were not actions and did not involve an agent. In VOL. 14, NO. 6, NOVEMBER 2003

PSYCHOLOGICAL SCIENCE

A. Wohlschläger et al. each condition, the action that had to be judged was a lever press followed by a pure tone. This sequence was selected for two reasons: First, making the action operant by including an action effect would encourage subjects to conceive their and the experimenter’s movements as involving agency, rather than reaction. Second, this common action effect was designed to emphasize the similarity between the three actions, notwithstanding that they involved different agents. We used planned contrasts to distinguish between the private and the integrationist views. The private view predicts that one’s own intentions play a key role in awareness of action. Therefore, according to this account, the perceived onset time of subjects’ own action would differ from the perceived onset times in the other two conditions, which would be equal. The integrationist view predicts that awareness of one’s own voluntary action and of observed actions should be comparable, because these activate a common neural system, but that awareness of these actions should be different from awareness of mechanical events. Therefore, according to the integrationist view, the perceived onset time for the subject’s own actions would be similar to the perceived onset time for another person’s actions, but both would differ from the perceived onset time of machine-generated actions.

Method Twelve subjects, 9 women and 3 men, all right-handed and between 21 and 47 years old, were tested in the apparatus schematically shown in Figure 1. Subjects viewed a response lever. Depressing the response lever 2 mm closed an electrical contact. An image of a clock was projected onto the table next to the lever. The clock had a single hand, 1.2 cm long, that rotated with a period of 2,560 ms within a face marked at conventional 5-“minute” intervals (Libet et al., 1983). The experimenter initiated the rotation of the clock at the start of each trial. The initial position of the

clock hand was random. The hand then rotated until the response lever was moved (see the descriptions of the three conditions later in this section), continued for a random period between 1.5 and 2.5 s thereafter, and then stopped. The subject then verbally reported the position of the clock hand when the response lever was depressed. In the self-action and machine-action conditions, the experimenter sat at a table about 2 m away from the experimental apparatus (in the other-action condition, the experiment was closer, as described later). In these conditions, subjects could see the experimenter only if they raised and turned their head to the left. Subjects were encouraged to make their verbal reports as precise as possible, and to avoid confining themselves to the intervals marked on the clock face. The experimenter verified that the subject was looking at the clock face during each trial. Subjects performed 40 trials in each of three conditions. The conditions differed according to how the response lever was moved. In the self-action condition, subjects pressed the lever with the index finger of their right hand at a time of their own choice. They were instructed to avoid pressing at “obvious” clock positions (e.g., 0, 30) or in response to the onset of the hand’s rotation. These instructions were designed to give the subjects’ movements the quality of voluntary actions, rather than reactions. The subjects then judged the perceived onset time of their action using the clock. In the other-action condition, the subject observed the experimenter (now sitting next to the subject) pressing the response lever with his right index finger. The experimenter followed the same constraints as those in the self-action condition. He avoided making any subtle preparatory movements of his hand prior to the lever press itself, as such movements could warn the subjects of his impending movement. When the clock stopped, the subject judged the perceived onset time of the experimenter’s lever press. In the machine-action condition, the response lever was moved by a solenoid invisibly located at the lever’s fulcrum. Subjects were in-

Fig. 1. Apparatus used in the experiments. Subjects viewed the response lever through a semisilvered mirror (a). The mirror served to project the image of a clock on the computer screen next to the lever (b). VOL. 14, NO. 6, NOVEMBER 2003

587

PSYCHOLOGICAL SCIENCE

Perceived Onset Time of Actions structed that the lever would move automatically. In fact, the experimenter invisibly activated the solenoid at a different latency on each trial using a switch hidden under his table, out of subjects’ view. The order of conditions was completely counterbalanced across subjects. In all conditions, closure of the lever was followed after 250 ms by a short auditory stimulus (100-ms tone of 1 kHz, delivered over loudspeakers). The position of the clock hand at the time of the lever’s contact was recorded on a computer for off-line analysis. The temporal resolution of response times was 1 ms (as determined by the software package LabVIEW). We calculated a judgment error for each trial by comparing the position of the clock hand at lever closure with the subject’s judgment of when the lever moved. Negative judgment errors indicate anticipatory awareness of action; positive judgment errors indicate delayed awareness. A few trials were excluded from the analysis because the subject reported not paying attention to the clock or response lever during the trial. Mean judgment errors were used for statistical analysis.

Results and Discussion Mean judgment errors of the 12 subjects are shown in the top panel of Figure 2. Subjects had slightly delayed awareness of their own actions (5 ms) and of the experimenter’s actions (11 ms), but an anticipatory awareness of the machine’s displacement of the response lever (77 ms). We tested the integrationist hypothesis with a planned contrast in which we assigned contrast weights of 1 to the self-action and otheraction conditions and 2 to the machine-action condition. The integrationist hypothesis was strongly supported, t(11)  9.45, p  .000001. A second planned contrast tested the privacy hypothesis by assigning contrast weights of 2 to the self-action condition and 1 to the other conditions. This hypothesis was also supported, t(11)  3.98, p  .002. Because these two contrasts are nonorthogonal (r  .5), they are not mutually exclusive. However, comparison of the t statistics and inspection of the mean values (Fig. 2) shows that our data are more consistent with the integrationist hypothesis than with the privacy hypothesis. The same conclusion is supported by the coefficients of determination, r2  .89 for the integrationist hypothesis and r2  .59 for the privacy hypothesis. The results clearly show that the perceived onset time of one’s own actions is comparable to the perceived onset time of other people’s actions. Both are substantially later than the perceived onset time of a physically comparable machine event. The similarity between the voluntary-action and observed-action conditions deserves particular comment. The perceived onset times of self-generated and other-generated actions are quite similar, despite the very different sources of perceptual information about them. Information about one’s own voluntary actions includes at least two private sources of information, namely, proprioceptive feedback and efference copy. Of course, no such private information is available about the actions of other people. Experiment 1 provides clearer support for the integrationist view than for the privacy view. We suggest that conscious representations of action (a) may derive from the common neural system for action generation and action understanding (Gallese et al., 1996), (b) are independent of the agent who makes the action, and (c) are distinct from physical events that are not actions and are not associated with any agent. Experiment 1 showed a substantial difference between the machine-action condition and the other two conditions. Judgments were

588

Fig. 2. Mean error in judgment of onset time in the three conditions of Experiments 1, 2, and 3. Vertical lines indicate standard deviations of the judgment error. Negative judgment errors indicate anticipatory awareness of action; positive judgment errors indicate delayed awareness. most inaccurate in the machine-action condition (mean judgment error  89 ms). The direction of this difference was not predicted, and we return to this point in the General Discussion. Viviani and Stucchi VOL. 14, NO. 6, NOVEMBER 2003

PSYCHOLOGICAL SCIENCE

A. Wohlschläger et al. (1992) reported large differences in the spatiotemporal perception of more extended, continuous trajectories according to whether the movements were made by human agents or were mechanical in origin; perception of biological motion was more accurate. In our study, the kinematic displacement was minimal and without significant spatial pattern, yet a comparable difference was observed. Such differences may reflect either an “intentional stance” (Dennett, 1987) or a specialized module for the perception of biological actions (Stucchi & Viviani, 1993). The agency effect in Experiment 1 was confounded with a visual difference between conditions. In the self-action and other-action conditions, subjects saw a hand pressing the lever, whereas no hand was present in the machine-action condition. Experiment 2 removed this confound.

EXPERIMENT 2 Method The methods and conditions for Experiment 2 exactly replicated those of Experiment 1 except for the subject’s vision of the response lever. Subjects always saw a gloved hand on the response lever, so that visual inputs across conditions were balanced. In the self-action condition, they saw their own hand. In the other-action condition, they saw the experimenter’s hand, wearing an identical glove. In the machine-action condition, the subjects saw the experimenter placing a rubber hand, wearing an identical glove, on the response lever. The lever was inserted into a slit at the tip of the index finger of the glove, so that the index finger of the rubber hand would always move with the lever. The posture of the hand was made comparable across conditions, with the index finger extended to contact the response lever, and the other fingers and thumb held against the palm. The subjects’ view of the hand extended to a point just distal to the edge of the glove, and subjects never saw more proximal body parts. Twelve new subjects (10 women and 2 men, all right-handed, ages between 21 and 27) participated in Experiment 2.

Results and Discussion The mean judgment errors were 1 ms for the self-action condition, 5 ms for the other-action condition, and 19 ms for the machine-action condition (see Fig. 2, middle panel). The same planned contrasts were performed as for Experiment 1. The integrationist hypothesis was supported, t(11)  3.14, p  .0095. The privacy hypothesis was not supported, t(11)  0.73. In addition, we performed a post hoc comparison between the machine-action conditions of Experiments 1 and 2. This comparison showed a significant difference, t(22)  5.39, p  .0002. The pattern of effects found in Experiment 1 were clearly replicated, again confirming the integrationist position. Moreover, the privacy hypothesis was clearly not supported. Although Experiment 2 made the other- and machine-action conditions more similar, which should have favored the privacy hypothesis, it clearly could be rejected. Interestingly, awareness in the machine-action condition was less anticipatory in Experiment 2 than in Experiment 1. This reduced anticipation for machine displacement could reflect better balancing of visual inputs. Alternatively, the presence of a more realistic hand in the machine-action condition in Experiment 2 may have activated to some extent a system for understanding biological actions. Pavani, Spence, and Driver (2000) have suggested that subjects can attribute rubber hands VOL. 14, NO. 6, NOVEMBER 2003

to their own body as long as the postural arrangement of the hands is plausible. Moreover, the actions of anthropomorphic, but not nonanthropomorphic, machines can prime imitation by humans (Castiello, Lusher, Mari, Edwards, & Humphreys, 2002). We speculate that the realistic hand in this experiment partially evoked attributions of intentions. Other studies have shown that viewing another person’s body parts can activate schema-level representations of the same parts of one’s own body (Parsons, 1987, 1994; Reed & Farah, 1995). Similarly, one’s own hand position can influence judgments about visual hand stimuli (Sirigu & Duhamel, 2001). Future research might vary the anthropomorphic qualities of an artificial hand parametrically and investigate corresponding changes in awareness of action.

EXPERIMENT 3 Experiments 1 and 2 supported the hypothesis of a similar time course for the awareness of the actions of self and others. However, in both experiments, actions were followed by an auditory effect. In principle, differences in the perceived onset time of action between conditions could arise because of differential inference of agency. Indeed, Haggard et al. (2002) found that an action that produces an effect is perceived to occur later than an action that does not produce an effect. The results of Experiments 1 and 2 might be predicted if self- and othergenerated actions bind with the ensuing beep, but machine actions do not. Experiment 3 repeated the design of Experiment 1 without any beep after the lever press.

Method Except for the absence of the beep, the methods were identical to those in Experiment 1. Twelve new right-handed subjects (3 men and 9 women, ages 21–28) participated in Experiment 3.

Results and Discussion The mean judgment errors were 26 ms for the self-action condition, 9 ms for the other-action condition, and 57 ms for the machine-action condition (see Fig. 2, bottom panel). The same planned contrasts were performed as for Experiment 1. The integrationist hypothesis was supported, t(11)  4.12, p  .0017. The privacy hypothesis was not supported, t(11)  0.70. Hence, again, even without an auditory action effect that might have caused a selective temporal shift of awareness in the self-action and the other-action condition, the privacy hypothesis was rejected and the integrationist view was supported. We conclude that our results reflect a genuine difference in the time course of action awareness rather than differences between conditions in the strength of associations between actions and effects.

GENERAL DISCUSSION Our results point toward the surprising conclusion that the time course of people’s awareness of their own actions resembles the time course of their awareness of the actions of others. In three experiments, the integrationist hypothesis was strongly supported, whereas the privacy hypothesis received little or no support. This suggests that people’s conscious awareness of intentions and related mental states does not arise from a private source within their own minds. Traditionally, the prototype of private access to mental states has been somatic proprioception (Melzack & Wall, 1982; Wittgenstein, 1953). In this view,

589

PSYCHOLOGICAL SCIENCE

Perceived Onset Time of Actions conscious experience is necessarily restricted to the subject because subjects receive somatic information only from their own body and not from the bodies of others. Nevertheless, our results do not reflect this differential access. We conclude that private information plays little or no part in the aspects of action awareness we studied. Instead, our results suggest awareness of action distributes successfully across different agents. In this respect, action awareness resembles the mirror-neuron system reported in monkeys (Gallese et al., 1996) and its apparent analogue in humans (Fadiga et al., 1995; Rizzolatti, Fadiga, Matelli, et al., 1996). Our data suggest that the neural circuit underlying action generation and action understanding may also participate in constructing conscious experience. Interestingly, the same neural network has been implicated in other high-level functions, which typically require consciousness, such as communication (Rizzolatti & Arbib, 1998). Our results also raise the issue of whether people represent the actions of others by analogy with their own. According to this interpretation, one has private access to one’s own intentions, and these intentions generate conscious awareness of action (Haggard & Eimer, 1999). One can then infer the intentions of others from their actions, by analogy with one’s own private case. This account rescues the privacy view, but at the same time softens the concept of privacy, because other people can infer one’s “private” intentions from one’s behavior. Moreover, if awareness of action could arise from mental states inferred in other people, rather than directly experienced, then one’s awareness of one’s own action might likewise be inferential, and therefore not especially privileged. This analogical interpretation recalls Dennett’s (1987) concept of intentional stance. When one adopts an intentional stance, one necessarily attributes agency to another individual. Previous philosophical work has focused on quite high-level effects of intentional stance, such as the way people verbally describe the behavior of others. Our experiments provide quantitative data suggesting that much more fundamental aspects of perceptual awareness are affected. Gallese has similarly placed the ability to perceive the sensorimotor actions of others at the foundation of social understanding (Gallese & Goldman, 1998) and social empathy (Gallese, 2001). Work on the theory of mind (Baron-Cohen, 1995) has focused on how people represent the beliefs of others. Our data suggest that representing the intentional actions of others may follow similar principles. Indeed, in development, it seems likely that representation of others’ actions should precede representation of their beliefs. Finally, our experiments repeatedly showed that the “actions” of a nonbiological machine were perceived differently from the actions of true biological agents, even when the actions were visually identical (see also Shiffrar & Freyd, 1993; Stevens, Fonlupt, Shiffrar, & Decety 2000). In particular, the other-action and machine-action conditions differed only in the subject’s concept of how the action was generated. We suggest that subjects took the intentional stance in the other-action condition, but not in the machine-action condition. This clearly influenced their perception of the subsequent physical event. Castiello et al. (2002) found that observing biological actions influences with motor performance, whereas observing mechanical actions does not. Moreover, a study by Meltzoff (1995) confirmed that such effects depend on representing the mental states of others. Children correctly inferred the intention of an actor to separate two parts of an object even though he failed to achieve the intended goal: When the children imitated the actor, they immediately separated the two parts of the object. Conversely, when a machine demonstrated kinematically similar fumbling

590

movements that likewise failed to separate the two parts of the object, the children did not produce the target action. These data suggest that the concept of an agent influences motor performance. We suggest that the concept of agency also influences the construction of conscious experience (see Knoblich, 2002, for the role of action cues in self-recognition). One final feature of our data deserves additional comment. In three experiments, we found that the perceived onset time of a machine action differed significantly from the perceived onset time of biological actions. Moreover, we found that this difference consistently involved anticipatory awareness of machine actions, with subjects claiming that the machine actions occurred before they actually did. Although a difference between the machine-action and other-action conditions was predicted, the negative sign of the judgment error in the machine-action condition was not predicted, and may seem surprising. Given that subjects cannot have advance information about the movement of a machine, and appear not to attribute intentions to a machine, one might have expected a delayed awareness in the machine-action condition. Binding between actions and effects (Haggard et al., 2002) could explain the anticipation of machine actions in Experiments 1 and 2. Selfand other-generated actions are perceived shifted toward effects that they cause. Our results suggest machine actions are not perceived in this way. In Experiment 3, no beep occurred. Nevertheless, using the response lever inevitably produced a click. The click could have become the “effect” of the action, again allowing some intentional binding for the self- and other-action conditions, but not for the machine-action condition. However, this remains an ad hoc explanation: In future research, we plan to vary the presence of an effect more systematically. This may further clarify how and why the actions of biological agents are perceived differently from the actions of nonagents. We speculate that machine actions may be perceived differently from self- and other-generated actions because machine actions are not perceived with reference to the effects they cause. In the present experiments, we studied the perceived onset times of actions. Awareness of action clearly involves, in addition to subjective time, dimensions such as force, effort, and pain. However, the proprioceptive system ensures that one has privileged access to these other dimensions. Time, in contrast, is a common metric dimension that can be compared across all events. Our results show that differential activation of the proprioceptive system in self-generated and observed actions may mask the operation of a common mental process that integrates these two classes of action. Moreover, this common integrating process operates at levels of the human mind high enough to enter conscious awareness.

REFERENCES Aziz-Zadeh, L., Maeda, F., Zaidel, E., Mazziotta, J., & Iacoboni, M. (2002). Lateralization in motor facilitation during action observation. Experimental Brain Research, 144, 127–131. Baron-Cohen, S. (1995). Mindblindness: An essay on autism and theory of mind. Cambridge, MA: MIT Press. Castiello, U., Lusher, D., Mari, M., Edwards, M., & Humphreys, G.W. (2002). Observing a human or a robotic hand grasping an object: Differential motor priming effects. In W. Prinz & B. Hommel (Eds.), Attention and performance: Vol. XIX. Common mechanisms in perception and action (pp. 315–333). Oxford, England: Oxford University Press. Craighero, L., Fadiga, L., Rizzolatti, G., & Umiltà, C. (1999). Action for perception: A motor-visual attentional effect. Journal of Experimental Psychology: Human Perception and Performance, 25, 1673–1692. Craighero, L., Fadiga, L., Umiltà, C.A., & Rizzolatti, G. (1996). Evidence for visuomotor priming effect. NeuroReport, 8, 347–349.

VOL. 14, NO. 6, NOVEMBER 2003

PSYCHOLOGICAL SCIENCE

A. Wohlschläger et al. Dennett, D.C. (1987). The intentional stance. Cambridge, MA: MIT Press. Fadiga, L., Fogassi, L., Pavesi, G., & Rizzolatti, G. (1995). Motor facilitation during action observation—a magnetic stimulation study. Journal of Neurophysiology, 73, 2608–2611. Fried, I., Katz, A., McCarthy, G., Sass, K.J., Williamson, P., Spencer, S.S., & Spencer, D.D. (1991). Functional-organisation of human supplementary motor cortex studied by electrical-stimulation. Journal of Neuroscience, 11, 3656–3666. Gallese, V. (2000). The inner sense of action: Agency and motor representations. Journal of Consciousness Studies, 7(10), 23–40. Gallese, V. (2001). The ‘shared manifold’ hypothesis. Journal of Consciousness Studies, 8(5–7), 33–50. Gallese, V., Fadiga, L., Fogassi, L., & Rizzolatti, G. (1996). Action recognition in the premotor cortex. Brain, 119, 593–609. Gallese, V., & Goldman, A. (1998). Mirror neurons and the simulation theory of mindreading. Trends in Cognitive Sciences, 2, 493–501. Goldberg, G., Kwan, H.C., Borrett, D., & Murphy, J.T. (1984). Topography of the movement-associated scalp potential (masp) suggests initiation of spontaneous movement by the supplementary motor area. Archives of Physical Medicine and Rehabilitation, 65(10), 662–663. Haggard, P., Aschersleben, G., Gehrke, J., & Prinz, W. (2002). Action, binding and awareness. In W. Prinz & B. Hommel (Eds.), Attention and performance: Vol. XIX. Common mechanisms in perception and action (pp. 266–285). Oxford, England: Oxford University Press. Haggard, P., & Eimer, M. (1999). On the relation between brain potentials and the awareness of voluntary movements. Experimental Brain Research, 126(1), 128–133. Knoblich, G. (2002). Self-recognition: Body and action. Trends in Cognitive Sciences, 6, 447–449. Kornhuber, H.H., & Deecke, L. (1965). Hirnpotentialänderungen bei Willkürbewegungen und passiven Bewegungen des Menschen: Bereitschaftspotential und reafferente Potentiale. Pflügers Archiv, 284, 1–17. Libet, B., Gleason, C.A., Wright, E.W., & Pearl, D.K. (1983). Time of conscious intention to act in relation to onset of cerebral activity (readiness-potential): The unconscious initiation of a freely voluntary act. Brain, 106, 623–642. Maeda, F., Kleiner-Fisman, G., & Pascual-Leone, A. (2002). Motor facilitation while observing hand actions: Specificity of the effect and role of observer’s orientation. Journal of Neurophysiology, 87, 1329–1335. Meltzoff, A.N. (1995). Understanding the intentions of others: Re-enactment of intended acts by 18-month-old children. Developmental Psychology, 31, 838–850.

VOL. 14, NO. 6, NOVEMBER 2003

Melzack, R., &Wall, P. (1982). The challenge of pain. Harmondsworth, England: Penguin. Parsons, L.M. (1987). Imagined spatial transformations of one’s hands and feet. Cognitive Psychology, 19(2), 178–241. Parsons, L.M. (1994). Temporal and kinematic properties of motor behavior reflected in mentally simulated action. Journal of Experimental Psychology: Human Perception and Performance, 20, 709–730. Pavani, F., Spence, C., & Driver, J. (2000). Visual capture of touch: Out-of-the-body experiences with rubber gloves. Psychological Science, 11, 353–359. Reed, C.L., & Farah, M.J. (1995). The psychological reality of the body schema: A test with normal participants. Journal of Experimental Psychology: Human Perception and Performance, 21, 334–343. Rizzolatti, G., & Arbib, M A. (1998). Language within our grasp. Trends in Neurosciences, 21(5), 188–194. Rizzolatti, G., Fadiga, L., Gallese, V., & Fogassi, L. (1996). Premotor cortex and the recognition of motor actions. Cognitive Brain Research, 3, 131–141. Rizzolatti, G., Fadiga, L., Matelli, M., Bettinardi, V., Paulesu, E., Perani, D., & Fazio, F. (1996). Localization of grasp representations in humans by PET: 1. Observation versus execution. Experimental Brain Research, 111, 246–252. Shiffrar, M., & Freyd, J.J. (1993). Timing and apparent motion path choice with humanbody photographs. Psychological Science, 4, 379–384. Sirigu, A., & Duhamel, J.R. (2001). Motor and visual imagery as two complementary but neurally dissociable mental processes. Journal of Cognitive Neuroscience, 13, 910–919. Stevens, J.A., Fonlupt, P., Shiffrar, M., & Decety, J. (2000). New aspects of motion perception: Selective neural encoding of apparent human movements. NeuroReport, 11, 109–115. Stucchi, N., & Viviani, P. (1993). Cerebral dominance and asynchrony between bimanual two-dimensional movements. Journal of Experimental Psychology: Human Perception and Performance, 19, 1200–1220. Tanji, J. (2001). Sequential organization of multiple movements: Involvement of cortical motor areas. Annual Review of Neuroscience, 24, 631–651. Viviani, P., & Stucchi, N. (1992). Biological movements look uniform: Evidence of motorperceptual interactions. Journal of Experimental Psychology: Human Perception and Performance, 18, 603–623. Wittgenstein, L. (1953). Philosophical investigations. Oxford, England: Blackwell.

(RECEIVED 3/26/02; REVISION ACCEPTED 12/13/02)

591