Infants' understanding of object-directed action

When and in what ways do infants recognize humans as intentional actors? An important aspect of this larger question concerns when infants recognize specific human actions (e.g. a reach) as object-directed (i.e. as acting toward goal-objects). In two studies using a visual habituation technique, 12-month-old infants were tested to assess their recognition that an adult's reach is directed toward its target object. Infants in the experimental condition were habituated to a display in which an actor reached over a wall-like barrier with an arcing arm movement, to pick up a ball. After habituation infants saw two test displays, for which the barrier was removed. In the direct test event the actor reached directly for the ball, the arm tracing a visually new path, but the action consistent with attempting to reach for the object as directly as possible. In the indirect test event the actor traced the old path, reaching over in an arc, even though the wall was no longer present. This arm movement was identical to that in habituation but no longer displayed a reach going directly to its object. In a control condition infants saw the same movements but in a situation with no goal-object. In the experimental conditions, with a goal object present, infants looked longer at the indirect test event in comparison to the direct test event. In the control conditions infants looked equally at both indirect and direct test events. We conclude that sensitivity to human object-directed action is established by 12-month-olds and compare these results to recent findings by [Gergely, G., Nadasdy, Z., Csibra, G., & Biro S. (1995). Taking the intentional stance at 12 months of age. Cognition, 56, 165-193] and [Woodward, A. (1998). Infants selectively encode the goal object of an actor's reach. Cognition, 69, 1-34].     

Perception and production of object-related grasping in 6-month-olds

In this study, 6-month-olds’ perception of an object-related human grasping action was compared with their level of grasping performance using a within-participants design. In the action perception task, infants were presented with the video of an actor’s grasping movement toward an occluded target object. Subsequently, an expected and an unexpected final state of this grasping movement were presented simultaneously, and infants’ looking times were measured. In the action production task, infants were presented with three graspable objects. Infants’ grasping behavior was coded to be either palmar or thumb-opposite grasping. Results indicate that infants who were already able to perform a thumb-opposite grasp differentiated between the two final states in the action perception task by looking longer toward the unexpected final state. In contrast, infants who showed only palmar grasps looked equally long toward both final states. This finding supports the assumption that action perception and action control are already closely related in infants as young as 6 months.     

Value learning modulates goal-directed actions

With experience, particular objects can predict good or bad outcomes. This alters our perceptual response to them: Reliable predictors of salient outcomes are recognized faster and better than unreliable predictors, regardless of the value (gain, loss) of the outcome they predict. When attentional resources are constrained, learned value associations matter, causing recognition of gain-associated objects to be spared. Here, we ask how learned predictiveness and value change the way we interact with potentially rewarding objects. After associating virtual objects (drinking flutes) with monetary gains or losses, reaching for and grasping corresponding real objects depended on the object's learned value. Action was faster when directed at objects that previously predicted outcomes more rather than less consistently, regardless of value. Conversely, reaches were more direct for gain- than for loss-associated objects, regardless of their predictiveness. Action monitoring thus reveals how value learning components become accessible during action.     

Habit outweighs planning in grasp selection for object manipulation

Object-directed grasping movements are adapted to intended interactions with an object. We address whether adjusting the grasp for object manipulation is controlled habitually, based on past experiences, or by goal-directed planning, based on an evaluation of the expected action outcomes. Therefore, we asked participants to grasp and rotate a dial. In such tasks, participants typically grasp the dial with an excursed, uncomfortable arm posture, which then allows to complete the dial rotation in a comfortable end-state. We extended this task by manipulating the contingency between the orientation of the grasp and the resulting end-state of the arm. A one-step (control) group rotated the dial to a single target. A two-step group rotated the dial to an initial target and then in the opposite direction. A three-step group rotated the dial to the initial target, then in the opposite direction, and then back to the initial target. During practice, the two-step and three-step groups reduced the excursion of their grasps, thus avoiding overly excursed arm postures after the second rotation. When the two-step and three-step groups were asked to execute one-step rotations, their grasps resembled those that were acquired during the two-step and three-step rotations, respectively. However, the carry-over was not complete. This suggests that adjusting grasps for forthcoming object manipulations is controlled by a mixture of habitual and goal-directed processes. In the present experiment, the former contributed approximately twice as much to grasp selection than the latter.     

Components of visual prior entry

The prior entry hypothesis contends that attention accelerates sensory processing, shortening the time to perception. Typical observations supporting the hypothesis may be explained equally well by response biases, changes in decision criteria, or sensory facilitation. In a series of experiments conducted to discriminate among the potential mechanisms, observers judged the simultaneity or temporal order of two stimuli, to one of which attention was oriented by exogenous, endogenous, gaze-directed, or multiple exogenous cues. The results suggest that prior entry effects are primarily caused by sensory facilitation and attentional modifications of the decision mechanism, with only a small part possibly due to an attention-dependent sensory acceleration.     

Readers' experiences of characters' goals and actions

In this article, the authors examined readers' sensitivity to the match between characters' goals and characters' actions. In Experiment 1, readers integrated actions consistent with characters' goals more easily when there was a match between the extremeness of the actions and the urgency of the goals. In Experiments 2 and 3, characters' actions were consistent with either explicit or implicit goals. Participants showed different sensitivity to the mismatch between actions and urgent goals when they simply read the actions (Experiment 2) versus when they judged the likelihood of the actions (Experiment 3). Taken together, these results offer an account of how readers experience actions and goals when engaged in both local and global processing.     

Bias and learning in temporal binding: Intervals between actions and outcomes are compressed by prior bias

It has consistently been shown that agents judge the intervals between their actions and outcomes as compressed in time, an effect named intentional binding. In the present work, we investigated whether this effect is result of prior bias volunteers have about the timing of the consequences of their actions, or if it is due to learning that occurs during the experimental session. Volunteers made temporal estimates of the interval between their action and target onset (Action conditions), or between two events (No-Action conditions). Our results show that temporal estimates become shorter throughout each experimental block in both conditions. Moreover, we found that observers judged intervals between action and outcomes as shorter even in very early trials of each block. To quantify the decrease of temporal judgments in experimental blocks, exponential functions were fitted to participants’ temporal judgments. The fitted parameters suggest that observers had different prior biases as to intervals between events in which action was involved. These findings suggest that prior bias might play a more important role in this effect than calibration-type learning processes.     

The role of prior experience in language acquisition

Learners exposed to an artificial language recognize its abstract structural regularities when instantiated in a novel vocabulary (e.g., Gómez, Gerken, & Schvaneveldt, 2000; Tunney & Altmann, 2001). We asked whether such sensitivity accelerates subsequent learning, and enables acquisition of more complex structure. In Experiment 1, pre-exposure to a category-induction language of the form aX bY sped subsequent learning when the language is instantiated in a different vocabulary. In Experiment 2, while naíve learners did not acquire an acX bcY language, in which aX and bY co-occurrence regularities were separated by a c-element, prior experience with an aX bY language provided some benefit. In Experiment 3 we replicated this finding with a 24-hour delay between learning phases, and controlled for prior experience with the aX bY language's prosodic and phonological characteristics. These findings suggest that learners, and the structure they can acquire, change as a function of experience.     

Infant's perception of goal-directed actions performed by a mechanical device

The present work investigated whether by the end of the first year, infants interpret actions performed by a mechanical device as goal-directed and why they would do so. Using a modified version of the Woodward (1998) habituation paradigm, 9- and 12-month-old infants were tested in a condition in which they saw a mechanical claw performing an action (Study 1). When infants viewed the claw grasping and transporting objects to the back of a stage, 12-month-old but not 9-month-old infants interpreted the action as goal-directed. In Study 2, 9-month-olds received prior to habituation an information phase showing infants how a human held and operated the claw. This enrichment of infants’ knowledge enabled 9-month-old infants to interpret the action display as goal-directed. The role of the developing means-end understanding and tool-use for infants’ interpretation of actions performed by a mechanical device is discussed.     

Effect of emotions on the choice of goals and actions

The present paper concerns the influence negative emotions exert on motivational processes, i.e., on the selection of goals and corresponding actions. It is assumed that, in a negative emotional state, the goal to overcome this negative mood state takes precedence over the goal to succeed in a nonemotional task. Correspondingly, emotion-related activities should be preferred to task-related ones. In order to test this assumption, subjects in a negative and a neutral emotional state were given a series of activities, some suited for coping, some suited for task fulfilment. They were instructed to name (Experiment I) or to carry out (Experiment II) those activities they preferred at the very moment and to give a short reason for each choice. Results showed that subjects in a negative mood compared to those of the neutral control group were less occupied with the nonemotional task, but instead more frequently chose emotion-related activities with the explicit reason that these activities served to overcome their negative emotions.     

The perception of nested affordances: An examination of expert climbers

Nested affordances in climbing comprise of multiple sequential actions, encompassing more than isolated reaching and grasping. This study examined the extent to which the perception of (multiple) nested affordances in climbing can be understood relative to body-scaled anthropometrics (arm span) and action-scaled (maximal action capabilities) measures, in relation to sequential task-goals of varying complexity. The utilisation of functionally equivalent actions during the actualisation of nested affordances was also investigated. Participants were required to estimate the maximal distance a handhold could be reached in four conditions of varying complexity: nested affordances (reach to Touch; reach to Grasp) vs. multiple nested affordances (reach to grasp with one hand followed by Removing the other hand from the starting handhold; reach to grasp with one hand followed by Moving up to another handhold). Ten expert climbers first had to estimate their maximal reaching distance in these different conditions (Touch, Grasp, Remove, Move Up) before performing the climbing movement. This sequence of estimation-action was repeated four times in each condition. Results revealed that action-scaling measures better captured nested affordances when multiple sequential behaviours are nested than body-scaled measures. Our findings also suggest that expert climbers utilise functionally equivalent movements during the realisation of multiple nested affordances (Remove and Move Up conditions).     

Twelve- and 18-month-olds copy actions in terms of goals

In the context of an imitation game, 12- and 18-month-old infants saw an adult do such things as make a toy mouse hop across a mat (with sound effects). In one condition (House), the adult ended by placing the mouse in a toy house, whereas in another condition (No House) there was no house present at the final location. Infants at both ages usually simply put the mouse in the house (ignoring the hopping motion and sound effects) in the House condition, presumably because they interpreted the adult's action in terms of this final goal and so ignored the behavioral means. In contrast, infants copied the adult's action (both the hopping motion and the sound effects) when no house was present, presumably because here infants saw the action itself as the adult's only goal. From very early, infants' social learning is flexible: infants focus on and copy either the end or the means of an adult action as required by the context.     

Infants attribute goals even to biomechanically impossible actions

Human infants readily interpret the actions of others in terms of goals, but the origins of this important cognitive skill are keenly debated. We tested whether infants recognize others′ actions as goal-directed on the basis of their experience with carrying out and observing goal-directed actions, or whether their perception of a goal-directed action is based on the recognition of a specific event structure. Counterintuitively, but consistent with our prediction, we observed that infants appear to extend goal attribution even to biomechanically impossible actions so long as they are physically efficient, indicating that the notion of ‘goal′ is unlikely to be derived directly from infants′ experience.     

Moral responsibility for actions: epistemic and freedom conditions

I will argue that a desert-based justification for treating a person in a certain way is a justification that holds this treatment to be justified simply by what the person is like and what he or she has done, independent of (1) the fact that treating the person in this way will have good effects (or that treating people like him or her in this way will have such effects); (2) the fact that this treatment is called for by some (justified) institution or practice; or (3) the fact that the person could have avoided being subject to this treatment by choosing appropriately, and therefore cannot complain of it. I will explore the implications of this understanding of desert for the role of desert-based justifications of blame, punishment, and economic reward.     

Contexts of a person's prior intentions facilitate observational learning in 2.5-year-old children

We examined whether contexts suggesting an actor's prior intentions facilitate observational learning in 2.5-year-olds. In Experiment 1, children observed an experimenter handle one box before proceeding to open a second box. In two prior intention conditions, children either watched the experimenter extract a toy from the first box or saw that the box had already been opened. In two no prior intention conditions, children watched the demonstration with only the second box or paired with irrelevant actions upon the first box. Children successfully opened the second box more often in the two prior intention conditions than in the two no prior intention conditions. Experiment 2 investigated stimulus generalization as another explanation for these results. A functionally different trap-tube task served as the pre-demonstration apparatus. Before watching the experimenter open the box, children either saw her extract a toy from the tube with a stick or observed the toy accidentally fall from the opening. In both cases, children opened the box at similar high rates. We discuss children's use of others’ prior intentions or observable outcomes in observational learning.     

Beyond the information given: Infants’ transfer of actions learned through imitation

Five experiments were conducted to investigate infants’ ability to transfer actions learned via imitation to new objects and to examine what components of the original context are critical to such transfer. Infants of 15 months observed an experimenter perform an action with one or two toys and then were offered a novel toy that was not demonstrated for them. In all experiments, infants performed target actions with the novel toy more frequently than infants who were offered the same toy but had seen no prior demonstrations. Infants exhibited transfer even when the specific part to be manipulated looked different across the toys, even when they had not performed the actions with the demonstration toys themselves, even when the actions produced no effects on the demonstrations, and even when the actions were demonstrated with only a single exemplar toy. Transfer was especially robust when infants not only observed but also practiced the target actions on the demonstration trials. Learning action affordances (“means”) seems to be a central aspect of human imitation, and the propensity to apply these learned action affordances in new object contexts may be an important basis for technological innovation and invention.     

Infants' understanding of actions performed by mechanical devices

Recent research suggests that 9-month-old infants tested in a modified version of the A-not-B search task covertly imitate actions performed by the experimenter. The current study examines whether infants also simulate actions performed by mechanical devices, and whether this varies with whether or not they have been familiarized with the devices and their function. In Experiment 1, infants observed hiding and retrieving actions performed by a pair of mechanical claws on the A-trials, and then searched for the hidden toy on the B-trial. In Experiment 2, infants were first familiarized with the experimenter and the claws but not their function. In Experiment 3, infants were familiarized with the function of the claws. The results revealed that search errors were at chance levels in Experiments 1 and 2, but a significant proportion of the infants showed the A-not-B error in Experiment 3. These results suggest that 9-month-old infants are less likely to simulate observed actions performed by mechanical devices than by human agents, unless they are familiarized with the function of the devices so that their actions are perceived as goal-directed.     

Reasoning about actions and obligations in first-order logic

We describe a new way in which theories about the deontic status of actions can be represented in terms of the standard two-sorted first-order extensional predicate calculus. Some of the resulting formal theories are easy to implement in Prolog; one prototype implementation—R. M. Lee's deontic expert shell DX—is briefly described.This research was partially supported by the Esprit III Basic Research Working Group No. 8319 ModelAge.     

Change in children's understanding of others' intentional actions

The study investigated children's intention understanding using a longitudinal design. Thirty-two Italian children were tested on the 'Demonstration of Intention' in the Re-enactment paradigm devised by Meltzoff (1995a), at two ages. Mean age was 12 months at the first session and 15 months at the second session. Previous research by Meltzoff showed that 18-month-olds are able to re-enact what the adult intended to do, when an adult tried but failed to perform certain target acts. Bellagamba and Tomasello (1999) found that, compared to 18-month-olds, 12-month-olds are less able to imitate unsuccessful goal-directed actions. The current study had two main goals. The first was to replicate earlier findings on intention understanding in 12- and 15-month-olds longitudinally assessed. The second goal was to analyse which actions children perform when not producing the target act. A coding scheme was devised to evaluate these non-target actions. Results confirmed previous findings, i.e. children's tendency to produce the target act was higher at an older age. At a younger age the most frequent response was 'undifferentiated manipulation' of target objects, followed by 'approximation to the target action'. Since children differed in their intention understanding performance at the first session, we decided to check whether these individual differences remained stable at the second session. Individual children with high intention understanding performance at 12 months tended to remain high at 15 months of age.     

Enacted and nonenacted encoding of social actions

Previous research has demonstrated that recall of enacted verbal commands is superior to recall of the same commands without enactment. The experiment reported explored whether the same effect would hold true in a social context as opposed to the non-social context used in previous research. The results demonstrated that this is indeed the case. Enacted verbal commands are better recalled than commands that are encoded verbally and better than commands that are encoded by means of observing them being performed. It was also demonstrated that items rated as having positive emotional value were better recalled than items rated as negative or neutral. This was true for enacted as well as for nonenacted commands. It is concluded that there is no basic difference between memory of commands enacted in the social context and memory of commands enacted in a non-social context. We discuss the data in relation to current theory of memory of actions.