Abstract
Human motion perception is crucial for social interactions. There is evidence that this perception is influenced by the knowledge of our body and its range of movement. We run two experiments to understand how robust this influence can be. First, we compared human and clock hand motion perception through an apparent motion paradigm. Second, we used a masked priming paradigm to explore how unconscious processes affect motion perception. While the clock hand rotations were generally perceived as clockwise, the human hands were perceived as rotating clockwise and counterclockwise, and their perception was predominantly aligned with biomechanical constraints. The main finding was that this alignment persisted under visual priming for human hands but not for clock hands. The priming effect was significantly reduced when the primed direction conflicted with biomechanically possible hand movements. This suggests that body knowledge shapes motion perception, with this effect proving highly robust.
Similar content being viewed by others
Introduction
The ability to perceive human motion is crucial for navigating a dynamic social world and successfully interacting with others. Numerous studies have proved this ability in cases where visual information is sparse or ambiguous1. Strikingly, motion perception has been demonstrated to occur even when movements are apparent rather than real2,3,4,5,6,7,8.
In a pioneering series of experiments, Shiffrar and Freyd presented participants with two rapidly flashing photographs depicting different body postures5,6. The postures were selected to suggest the initial and final steps of actions, such as arm, leg, or head rotations. Participants perceived movement trajectories adherent to the biomechanical constraints of the involved effectors when the temporal rate of photograph presentation matched the actual execution of the suggested action. However, when the temporal rate of photograph presentation was considerably faster than the movement execution, participants also perceived biomechanically implausible movement trajectories.
Subsequent research has corroborated the claim that body biomechanics constrains motion perception. Chatterjee et al.2 found that this effect applies not only to movements of the human body but also to movements of nonbiological objects that match the length, orientation, and position of a moving limb. Orgs et al.4further demonstrated that even when irrelevant to the task, our perception of bodily postures may also impact our ability to detect human motion (see also9). Finally, Vandenberghe and Vannuscorps10showed that biomechanical constraints critically affect how people perceptually extrapolate bodily movements when required to individuate the position where a disappeared bodily effector would have been if it had briefly continued to move (see also11,12).
Several studies on mental imagery have also described the biomechanical constraints influencing motion perception13,14,15,16,17. For example, participants were faster in mentally imagining rotating their right hand when the starting position of their arm was biomechanically plausible compared to when it violates joint constraints15. Similarly, in a mental rotation task requiring grasping a bar with a hand, participants’ response time followed a typical mental rotation function when the movement sequences were biomechanically consistent, exhibiting a monotonically increasing trend with rising angular bar disparity. However, this was not the case for biomechanically implausible movement sequences, which were uniformly slow16.
The biomechanical constraint effect is posited to rely on the knowledge of our own body, encompassing the relations of its parts and their range of movement12,18,19,20,21. This knowledge can gather information from diverse sources, including visual, tactile, proprioceptive, and motor inputs, and influences how we perceive our own and others bodily movements22,23,24.
Despite progress in studying the biomechanical constraint effect, more research is needed to probe the robustness of the influence exerted by the knowledge of our body on motion perception. Previous research suggested that motion perception can also be biased by our knowledge of objects and their usual range of movement25,26,27. When an object in motion comes to a sudden stop, people are usually biased in representing the final position it would have reached if the motion had continued28. The magnitude of this bias, often referred to as “representational momentum,” is influenced by knowledge of an object’s typical motion. Indeed, Reed & Vinson25showed that knowing that an upward-moving ambiguous stimulus was a NASA rocket induced a larger representational momentum effect than knowing that it was a steeple. Similar results have been obtained by measuring the impact of knowledge of objects on estimating their speed27.
A natural question arises as to whether knowledge of our body diverges from knowledge of objects in shaping motion perception. Does the former similarly influence motion perception as the latter? Or does body knowledge exert a more robust bias on motion perception than object knowledge? Answering these questions contributes to a better understanding of how body knowledge shapes motion perception. The main aim of the present study was to advance in this direction. To achieve this, we adopted a two-step strategy. First, along the lines of previous research, we systematically contrasted the influence of body and object knowledge on motion perception, considering whether external factors can modulate this influence (i.e., Experiment 1). Second, we leveraged this preliminary systematic investigation to explore whether unconscious processing of stimulus features can affect how body and object knowledge shape motion perception (i.e., Experiment 2).
In Experiment 1, we assessed the impact of body and object knowledge on motion perception by contrasting human and clock hands through an apparent motion paradigm. Participants viewed two static photographs displaying eight different initial and final positions of a human or a clock rotating hand, with all apparent rotations spanning 180 degrees (see Fig. 1 and “HAND” and “CLOCK” GIFs at the following link). The human hand was presented as either a right or left hand, while the clock hand was positioned on either the right or left of the face to mimic human hand positions. Participants were instructed to determine the direction of the apparent rotation (clockwise or counterclockwise) by pressing designated keys with their index and middle fingers on either their right or left hand.
Experimental paradigm of Experiment 1. Panel A shows the right human hand pairs employed in Experiment 1. Panel B shows an example of an experimental trial with human hand stimuli (45°−225° rotation, right hands). Panel C shows an example of an experimental trial with clock hand stimuli (45°−225° rotation, right hands). Each trial began with a 500 ms fixation cross, followed by the two static photographs (100 ms duration) separated by an inter-stimulus interval varying between 10, 50, 200, and 300 ms. Then, the fixation cross reappeared, and the subject responded, indicating the perceived direction of the rotation (clockwise – right arrow – or counterclockwise – left arrow). Participants responded with different hands in different blocks.
We expected body and object knowledge to modulate apparent motion perception in line with previous research. Human hands should exhibit clockwise and counterclockwise rotations as long as they do not violate body biomechanics. Conversely, participants should tend to perceive clock hands as rotating in a clockwise direction. Compared to previous works, our study not only tested these effects in a broader variety of rotations but also allowed us to verify possible modulations of such effects induced by the congruency of the perceived stimulus with the participants’ response hand (right or left). Indeed, there is evidence that the congruency between the perceived and the responding hand might impact participants’ performance in various tasks ranging from mental imagery29to cognitive control30.
In Experiment 2, we capitalized on the preliminary investigation of human and clock hand (apparent) motion perception to stride forward in probing the robustness of the perceptual biases induced by body and object knowledge. Specifically, we examined whether and to what extent these perceptual biases could be modulated by unconscious processing of stimulus features, mainly when they are expected to affect motion perception in ways that potentially conflict with the knowledge of how our body and a familiar object, such as a clock, move.
Prior research has demonstrated that visual priming influences the apparent motion perception of non-biological objects (i.e., dot patterns) by inducing preferential motion directions in animation sequences that would otherwise be ambiguous31,32,33,34,35,36,37. However, how visual priming affects apparent human motion perception remains substantially under-investigated (see3 for an exception). The main aim of our study is to fill this gap.
To this aim, we leveraged a masked priming paradigm developed by Dehaene et al.38, also called the “sandwich” paradigm, presenting a prime immediately preceded and followed by two mask images. This paradigm has the advantage of rendering the prime virtually invisible. It is well known that a stimulus can bias (prime) subsequent perception, even when it is not consciously perceived. This bias may exhibit a biphasic pattern. When the prime duration or the interval between the prime and target is very short, the prime positively biases the target, while a negative bias occurs when the prime duration or the interval is longer39,40,41.
We modified the apparent motion task used in Experiment 1 by selecting two rotations (45°−225° and 135°−315°). Both rotations involved the human right hand and the clock hand positioned on the right side of the clock face. According to Experiment 1, the first rotation (i.e., 45°−225°) was perceived as clockwise for both human and clock hands, while the second rotation (135°−315°) was perceived as ambiguous for both stimuli. We introduced a prime image consisting of a human or clock hand positioned in a way that suggests a clockwise or counterclockwise rotation. Such a prime was immediately preceded and followed by a scrambled mask (see Fig. 2), thus producing positive priming, suggesting either a clockwise or a counterclockwise rotation.
Experimental paradigm of Experiment 2. Panel A shows human and clock hand stimuli pairs employed in Experiment 2. Panel B shows an example of an experimental trial with human hand stimuli (45°−225° rotation). Panel C shows an example of an experimental trial with clock hand stimuli (45°−225° rotation). Each trial began with a 500 ms fixation cross, followed by the prime consisting of the forward mask (71 ms), the prime Fig. (43 ms), and the backward mask (71 ms). Then, the two static photographs (100 ms duration) appeared separated by a 300 ms inter-stimulus-interval. Then, the fixation cross reappeared, and the subject responded with the right hand, indicating the perceived direction of the rotation (clockwise – right arrow – or counterclockwise – left arrow).
This allowed the testing of two hypotheses. The first hypothesis proposes that visual priming will align motion perception with the primed direction when the apparent motion of both human and clock hands is ambiguous. The second hypothesis concerns the visual priming effect when the apparent motion has a preferential (e.g., clockwise) perceived direction. If body knowledge influences motion perception, we expect visual priming for human hands to be more effective for clockwise than counterclockwise motion. This asymmetry would occur because the counterclockwise direction conflicts with body biomechanical constraints, which only allow clockwise hand rotation. If object knowledge about clocks is as powerful as body knowledge, clock hands should also show stronger priming effects for clockwise versus counterclockwise motion, reflecting our strong associations with conventional clock movement. However, if object knowledge is weaker than body knowledge, clock hands should show equal priming effects in both directions, while human hands would still show the clockwise advantage. Such a pattern would provide evidence that body knowledge specifically shapes how we perceive motion, and that this effect is particularly robust compared to other types of knowledge.
Results
Experiment 1
In Experiment 1, we pursued two aims: to characterize perceptual differences between human and clock hand motion and to investigate whether responding with one hand (e.g., right) interferes with the biomechanical constraint effect when observing an incongruent hand (e.g., left). We collapsed the data across all ISI conditions because there were no evident differences between the various ISIs. We then averaged responses separately for left and right hand stimuli in each rotation direction, human and clock stimuli, and congruent and incongruent conditions. Since biomechanics constrains human hand movements in opposite directions for right and left hands, we flipped the left hand responses (i.e., when a clockwise response was prompted, we scored it as counterclockwise and vice versa) before merging. Otherwise, opposite responses to right and left human hands following biomechanical constraints would have artificially led to a 50% response (i.e., chance level). The percentage of clockwise responses in a 2 × 2 × 8 ANOVA with the following within-subject factors: STIMULUS (Clock, Human), CONGRUENCY (Congruent, Incongruent), and ROTATION (0°−180°, 45°−225°, 90°−270°, 135°−315°, 180°−0°, 225°−45°, 270°−90°, 315°−135°).
Besides a significant main effect of ROTATION (F7,119 = 33.93; p < 0.001; η2p = 0.67), importantly, the results highlighted a significant main effect of STIMULUS (F1,17 = 20.15; p < 0.001; η2p = 0.54; see Fig. 3A), with a significantly greater percentage of clockwise responses in the clock as compared to human hands. The significant main effect of STIMULUS supports our hypothesis that human and clock hands constrain motion perception in different ways. In line with this hypothesis, results also showed a significant STIMULUS × ROTATION interaction (F7,119 = 13.95; p < 0.001; η2p = 0.45; see Fig. 3B), suggesting a smaller percentage of clockwise responses for human than clock hand stimuli in those rotations where a counterclockwise direction was the only biomechanically plausible one.
The plot represents the results of Experiment 1, depicting the main effect of PRIME (Panel A) and the STIMULUS × ROTATION significant interaction (Panel B). Values represent the percentage of clockwise responses for human (violet) and clock (bronze) hand stimuli. Bars represent the standard error of the mean (SEM).
The three-way STIMULUS × CONGRUENCY × ROTATION interaction was not statistically significant (F7,119 = 0.83; p = 0.57; η2p = 0.046), indicating that the hypothesis of a congruency effect on how participants perceived human motion was not verified.
Experiment 2
To assess whether and to what extent the influence of body and object knowledge on motion perception could be modulated by the unconscious processing of stimulus features, we entered the percentage of clockwise responses in a 2 × 2x2 ANOVA (PRIME: Clockwise, Counterclockwise; STIMULUS: Clock, Hand; ROTATION: 45°−225°, 135°−315°).
The ANOVA revealed a significant main effect of PRIME (F1,20 = 51.91; p < 0.0001; η2p = 0.72), with a greater percentage of clockwise responses when the prime suggested a clockwise than a counterclockwise rotation. This effect supports the effectiveness of our priming task in influencing participants’ overall responses. Furthermore, significant STIMULUS × ROTATION (F1,20 = 4.92; p = 0.038; η2p = 0.2) and STIMULUS × PRIME (F1,20 = 18.41; p < 0.001; η2p = 0.48) interactions emerged. Crucially, also the three-way STIMULUS × ROTATION × PRIME interaction was significant (F1,20 = 22.47; p < 0.001; η2p = 0.53; see Fig. 4). In the 135°−315° rotation, human and clock hands stimuli elicited mean responses compatible with the primed direction. In the 45°−225° rotation, the percentage of clockwise responses was consistent with the prime when it suggested a clockwise direction for both human and clock hand stimuli. When the prime suggested a counterclockwise rotation, incompatible with biomechanical (body-related) constraints in the 45°−225° rotation, the percentage of clockwise responses approached 50% for human hand stimuli. In contrast, it followed the primed direction for clock hand ones. This interaction pattern suggests that when the prime and body knowledge bias participants towards the same direction, the effect will be large, while when they bias participants towards opposite directions, the effect will be small. This was not the case for object knowledge, with the prime bias being the same in both directions.
The plot represents the results of Experiment 2, depicting the STIMULUS × ROTATION × PRIME significant interaction. Values represent the percentage of clockwise responses to each stimulus for human (in pink) and clock (in brown) hand stimuli in clockwise (solid lines) and counterclockwise (dotted lines) primed conditions. Bars represent the standard error of the mean (SEM).
Finally, the responses in Experiment 2 were, on average, less biased than those in Experiment 1, especially for the 45°−225° rotation. This result could be attributed to the fact that, in Experiment 2, participants also perceived mask stimuli, which may have interfered with the biomechanical constraint effect on their performance. Another possibility is that participants wanted to select both responses equally across Experiment 2. However, the results of Experiment 2 show a significant interaction between prime, stimulus, and rotation, which cannot be fully explained by either of these interpretations.
Control experiment
We ran a control experiment to test the participants’ accuracy in consciously perceiving the prime image features. The same paradigm of Experiment 2 was presented again to the same experimental subjects,the study, participants sat at a desk facing a computer but, in this case, participants were informed about the existence of the prime stimulus, and they were asked to pay attention to it and to report whether it consisted of a rightward or leftward image in a two-alternative forced-choice task. Participants’ accuracy was compared to the chance level through a one-sample Wilcoxon Signed Rank Test, which showed no significant difference (W = 93, p = 0.06). Accordingly, participants exhibited responses to the control task around the chance level (mean accuracy ± SD = 53.12 ± 6.09).
Discussion
The current study sought to enhance our understanding of how body knowledge shapes motion perception.
In Experiment 1, we systematically investigated similarities and differences in perceiving the apparent rotations of clocks and human hands. In line with previous research2,4,5,6,7,8, we found differential biases in the perception of clock and human hand rotations. Our data did not support the hypothesis that incongruence between the used and observed hands would impact these biases in human motion perception. Compared to human hands, clock hands elicited, on average, significantly more clockwise motion perception (see Fig. 3A). The perception of clock hand rotations exhibited a bias toward the clockwise direction for all angles on average. In contrast, the human hand could be perceived as rotating in either a clockwise or counterclockwise direction, resulting in an averaged percentage of clockwise responses of around 50%. For instance, the perception of 0°−180° and 45°−225° right human hand rotations was biased towards the clockwise direction. Conversely, the 180°−0° and 225°−45° right human hand rotations were biased in the counterclockwise direction (see Fig. 3B). These preferential directions were aligned with body biomechanics. Indeed, the counterclockwise direction was biomechanically implausible for the 0°−180° and 45°−225° rotations, as well as the clockwise direction for the 180°−0° and 225°−45° rotations of the right human hand.
For some rotation pairs (90°−270°, 135°−315°), human and clock hands elicited similar perception patterns, with neither stimulus type displaying a discernible directional preference. In the case of human hand rotations, their ambiguity could be easily accounted for by appealing to body knowledge. In those rotations, the clockwise and the counterclockwise directions were biomechanically equally plausible. On the contrary, the observed ambiguity in clock hand rotations could not be immediately explained by their standard range of movement alone. Instead, this ambiguity may be partially attributed to the gravity effect—a known phenomenon where gravity-related knowledge biases motion perception19. This gravitational bias could explain the ambiguous perception of clock hand rotations, particularly since many involved upward trajectories. Additional factors may have contributed to the ambiguity of the perceived clock hand rotations. We depicted clocks as geometrical figures rather than photographs of actual clocks to control for low-level visual features between human and clock hand stimuli. While this design enhanced experimental control, it potentially decreased the clockwise bias compared to what would have been observed with more realistic clock hand representations. Furthermore, the alternating presentation of human and clock hand rotations may result in the human hand movements influencing the perception of clock hand rotations, though no reciprocal effect was observed.
The systematic investigation of the similarities and differences in the perception of the clock and human hand rotations enabled us to investigate the robustness of body and object knowledge by assessing the potential impact of unconscious perceptual processes. As mentioned in Experiment 1, we found that some human and clock hand rotations have a preferential direction, while other human and clock hand rotations do not. In particular, the 45°−225° rotation was perceived as having a clockwise direction for the clock and human hands. In contrast, the 135°−315° rotation did not clearly exhibit a similar preferential direction for the clock and human hands. In Experiment 2, we took advantage of this finding by priming both 45°−225° and 135°−315° rotations with a photograph depicting either a human or a clock hand oriented to suggest a clockwise or counterclockwise rotation direction. A control experiment ruled out the possibility that participants consciously elaborated the features of priming stimuli.
Our results revealed a differential impact of visual priming on the clock and human hands’ perceived rotations. Specifically, we observed that the perceived direction of both clock and human hands varied in response to clockwise or counterclockwise prime stimuli within the 135°−315° range. In contrast, a differential pattern between the clock and human hands was observed within the 45°−225° rotation range. The former result highlights that, when both rotations are possibly perceived, the prime modulates its perception along a direction in object stimuli (as shown by previous research31,32,33,34,35,36,37) and in body stimuli. Conversely, the latter result gives us an interesting insight into the robustness of how object and body knowledge shape motion perception. Indeed, in the 45°−225° rotation range, under the clockwise primed condition, both clock and human hands were consistently perceived as moving clockwise. In contrast, in the counterclockwise primed condition, the human hands were reported as moving less in the unnatural counterclockwise direction than the clock hands.
According to Experiment 1, in the 45°−225° rotation range, the clockwise primed direction aligns with object and body knowledge constraints, whereas the counterclockwise primed direction diverges from both. Our results pinpoint that, while the clock hand rotations were perceived according to the direction imposed by the visual priming, human hands showed some degree of resistance to the counterclockwise primed rotation, as this rotation is biomechanically implausible. Thus, our results advocate for a stronger influence of body than object knowledge. While the perception of clock hands always followed the direction suggested by the priming, in the case of the human hand, the effect of priming is significantly reduced when the suggested direction violates the biomechanical constraints that regulate body movements.
These results align with those of a previous study3in which two priming views of a rotating human body’s initial and final posture were interspersed with a targeted view of a novel human body posture biomechanically plausible or implausible. The results showed that the priming effect occurred only when the target view was biomechanically plausible. The novelty of our study is that participants were unaware of the priming stimuli’s features. In addition, we explicitly avoided any potential confounding factors related to critical differences in the static pictures. Indeed, unlike Kourtzi and Shiffrar3, the static images used in all the priming and primed stimuli did not individually represent biomechanically implausible body postures, nor did they differ regarding visual familiarity.
The finding that the perceived human hand rotations did not exhibit a counterclockwise direction, even when primed in this direction, suggests that the bias on motion perception induced by body knowledge was robust enough to counter the competing bias generated by the visual priming when the primed direction was explicitly in contrast with the knowledge of rotating hand biomechanics.
This does not apply to the knowledge of an object like a clock. The visual priming had the same impact on the perception of both clock hand rotations, equally biasing them in the clockwise and counterclockwise directions. Given the lack of physical constraints, one could be tempted to explain this effect by arguing that clock hands can rotate equally in both a clockwise and a counterclockwise direction. Furthermore, although the depicted clock hands prevented priming from low-level visual confounds, they were less familiar than those participants were accustomed to. In principle, one cannot exclude that both these factors have somehow influenced the outcome of visual priming on the clock hands. However, Experiment 1 showed that clock hands were generally perceived as rotating in a clockwise direction. This was particularly evident in the clock hand rotation (45°−225°) chosen for Experiment 2. Additionally, it is worth noting that in Experiment 1, even when bistable, clock hand rotations were not perceived as counterclockwise on average.
Overall, our results suggest that body knowledge differs from object knowledge in shaping motion perception. However, the differential impact of body and object knowledge on motion perception is not per se a cue for a difference in the mechanism involved. Some have hypothesized that the body knowledge’s impact on motion perception could be solely or primarily accounted for in terms of perceptual learning21,42,43. In contrast, others have proposed that motor information would play a critical role18,20.
Our study does not offer a substantial way of untangling the mechanisms specifically involved in body knowledge’s shaping motion perception, as this was beyond its scope. Instead, it provides insight into how robust the influence that the knowledge of our body and its range of movement exerts on the way we perceive the movements of others. The knowledge of our body appears to be distinct from the knowledge of objects. This beckons for future research to ascertain whether the bias arising from the body knowledge results from perceptual learning distinguished by its persistent and consistent nature12,43or involves the recruitment of motor processes and representations typically associated with action planning and execution44,45,46. Regardless of which account proves correct, it must explain why body knowledge fundamentally shapes motion perception and why this effect persists, at least partially, despite visual priming interference.
Methods
Participants
Experiment 1
This experiment involved 18 healthy participants, including 12 women. Their mean age was 26.5 years (SD = 3.20), and their mean years of education were 17.22 (SD = 1). The sample size was estimated using a priori analysis of G*Power software (www.psycho.uniduesseldorf.de/abteilungen/aap/gpower3) on a pilot dataset comprising the first 5 participants. Since the effect of interest was the difference between human and clock hand stimuli in each of the eight possible rotations, dz was calculated by comparing human and clock hand stimuli in the 45°−225° rotation, i.e., a rotation wherein a difference between stimuli was expected. A sample of 18 participants was estimated [α = 0.05/8; power (1 − β) = 0.98; dz = −1.3]. To corroborate these power analyses, we also applied a recently devised method by Anderson and colleagues47. This method returns the necessary sample size based on information obtained from previous publications or a previous pilot study (as in our case), correcting for publication bias or uncertainty. The estimated sample size was equal to 20 subjects (t-value of the comparison = 2.9; N = 5; alpha = 0.05/8; assurance = 0.5, power = 0.98, step = 001), confirming a similar size as the previous power analysis. All participants were right-handed, as assessed by the Edinburgh Handedness Inventory48, and had normal or corrected-to-normal visual acuity. They provided informed consent to participate in the study, which was approved by the Ethical Committee of the University of Turin (protocol n°122,571). All the approved experimental protocols were performed in accordance with the Declaration of Helsinki.
Experiment 2
This experiment involved another sample of 21 participants, including 15 women. Their mean age was 25.43 years (SD = 3.68), and their mean years of education were 15.38 (SD = 2.75). The sample size was estimated using a priori analysis of G*Power software (www.psycho.uniduesseldorf.de/abteilungen/aap/gpower3) on a pilot dataset comprising the first 5 participants. In Experiment 2, the effect of interest was the interaction between prime and stimulus. This was expected to result in significant differences between primed conditions in human and clock hand stimuli for the constrained rotation (i.e., 45°−225°). In contrast, differences in human and clock hand stimuli were not expected in the non-constrained rotation (i.e., 135°−315°). Hence, dz was calculated by comparing the delta between human and clock hand stimuli in clockwise and counterclockwise primed conditions in the 45°−225° rotation. A sample of 21 participants was estimated [α = 0.05/2; power (1 − β) = 0.98; dz = 1.03]. To corroborate these power analyses, we applied the method by Anderson and colleagues described above47, which returned a sample size equal to 19 subjects (t-value of the comparison = 2.6; N = 5; alpha = 0.05/2; assurance = 0.5, power = 0.98, step = 0.001), confirming the previous power analysis. All participants were right-handed, as assessed by the Edinburgh Handedness Inventory48, and had normal or corrected-to-normal visual acuity. They provided informed consent to participate in the study, which was approved by the Ethical Committee of the University of Turin (protocol n°122,571).
Study design and experimental procedures
Experiment 1
Experimental protocol. During the study, participants sat at a desk facing a computer screen 80 cm away. They rested their hands on the keyboard and completed an apparent motion task involving observing pairs of visual stimuli4,5,49. The stimulus pairs included two human hands or two clock hands. For the human hands, participants viewed two sequential photographs showing the starting and ending positions of eight wrist rotations performed with either the right or left hand (see Fig. 1). The clock hand stimuli consisted of two images showing the hands of a clock in different positions, creating apparent rotation trajectories that were equivalent to those of the biological stimuli. To match the biological stimuli, the clock hands were presented on either the right or left side of the clock face.
Participants were informed that the stimuli represented human and clock hands. They were asked to judge the direction of the perceived rotation (i.e., clockwise or counterclockwise) of human and clock hands by pressing either the right or left arrow key on a computer keyboard with their index or middle fingers, and participants’ responses were recorded as 1 (for counterclockwise rotations) or 2 (for clockwise rotations) (see Fig. 1B and 1 C). The apparent motion task was repeated in four different experimental blocks that resulted from the intersection between the response effector (right or left hand) and the observed stimuli (right or left observed human and clock hand). The following blocks were performed:
-
1.
The right human and clock hands were presented, and participants responded with their right hand (right hand congruent).
-
2.
The right human and clock hands were presented, and participants responded with their left hand (right hand incongruent).
-
3.
The left human and clock hands were presented, and participants responded with their left hand (left hand congruent).
-
4.
The left human and clock hands were presented, and participants responded with their right hand (left hand incongruent).
The order of the experimental blocks was counterbalanced among participants in terms of the observed stimulus side (right or left observed human and clock hand) and response side (right or left participants’ hand). Before the experimental blocks, participants completed two brief training blocks, in which we ensured that they could recognize the stimuli as human and clock hands, see the apparent motion, and provide the response correctly. During this familiarization phase, participants responded with either their right or left hand. The order of the training blocks was counterbalanced among participants so that half started with the right-hand training, and the other half started with the left-hand training.
Stimuli. The visual stimuli consisted of pairs of images depicting either human or clock hand rotations. Human hand images were shot with an actor (female, 170 cm, 21 years old) wearing a black t-shirt. The actor provided her consent to image release. For each rotation, the actor was required to assume the corresponding arm, wrist, and hand position. Multiple high-quality pictures were taken with the same light and distance position, using a Canon camera EOS 1000D. Then, the different hand positions were superimposed on a neutral arm and body position, and facial expressions and gender features were blurred to avoid distractions from the hand’s location. All the superimposed hand images come from real positions of the actual actor’s hand. Left stimuli were obtained by flipping the right stimuli images. Clock hand images were not real clock pictures but ad-hoc ones created to mimic the human ones as much as possible in terms of colors and position on the background to reduce possible low-level confounds. Human hands were shown from a third-person perspective and the backside. For both human and clock stimuli, there were eight different 180° rotations, and each pair of images showed the starting and end points of the rotation. The rotations were as follows: 0°−180°, 45°−225°, 90°−270°, 135°−315°, 180°−0°, 225°−45°, 270°−90°, and 315°−135° (see Fig. 1). Note that, for human hands, some of the observed rotations represented movements that could biologically be performed only in one direction (clockwise: 0°−180°, 45°−225°; counterclockwise: 0°−180°, 45°−225°), while some others could be performed in both directions (90°−270°, 135°−315°, 270°−90°, and 315°−135°). Each image was presented in the center of the screen for 100 ms, separated by an inter-stimulus interval (ISI) of 10, 50, 200, or 300 ms. A fixation cross appeared on the screen for 500 ms before each photograph pair. After the presentation of the visual stimulus pair, the fixation cross reappeared and remained until the participant responded. Participants were asked to gaze at the fixation cross throughout the experiment. Fixation was not experimentally controlled. Each rotation, stimulus type, and ISI combination was presented seven times for 448 trials.
Experiment 2
Experimental protocol. The experimental procedures of Experiment 2 were similar to those of Experiment 1. Participants underwent a single right congruent block of the apparent motion task described above, in which human and clock hand stimuli pairs were preceded by a forward- and backward-masked prime (see Fig. 2). The priming paradigm was adapted from Dehaene et al.38: the prime stimulus (43 ms) was preceded and followed by a mask (71 ms) consisting of a scramble of the prime image. The prime consisted of a human and a clock hand, suggesting a clockwise or a counterclockwise trajectory (see Fig. 2 and Figure S1 in the Supplementary Information). As in Experiment 1, participants were asked to judge the direction of the perceived rotation (i.e., clockwise or counterclockwise) of human and clock hands by pressing either the right or left arrow key on a computer keyboard with their right index or middle fingers and participants’ responses were recorded as 1 (for counterclockwise rotations) or 2 (for clockwise rotations). Before the experimental task, a training block was performed to ensure that participants could recognize the stimuli as human and clock hands, see the apparent motion, and provide the response correctly.
Stimuli. Human and clock hand stimuli consisted of two different rotations (45°−225°; 135°−315°), selected from those employed in Experiment 1. Among the possible rotations, we chose an (apparent) rotation whose direction was systematically reported as clockwise for both human and clock hands (45°−225°) and an (apparent) rotation whose direction was ambiguous for both human and clock hands (135°−315°). As in Experiment 1, each stimulus of the pair was presented in the middle of the computer screen for 100 ms and was divided by a 300 ms ISI. To prime a counterclockwise direction, we used images of human and clock hands at 45° and 135° angles; images of human and clock hands at 225° and 315° angles were used to prime clockwise rotations. The 135° and 315° images were employed as prime stimuli for the 45°−225° rotations, whereas the 45° and 225° images were the prime stimuli for the 135°−315° rotations. After the presentation of the visual stimulus pair, the fixation cross reappeared and remained on the screen until the participant responded. For each of the two rotations, stimulus types (human and clock hand), and priming (clockwise and counterclockwise), 28 trials were presented for a total of 224. The training block consisted of 16 trials of the experimental block.
Data analysis
Statistical analyses were performed using Jamovi Software (v2.3.28). Below is a detailed description of the data analyses conducted for each experiment.
Experiment 1
Experiment 1 addressed differences between human and clock hand stimuli across rotations (1) and the possible impact of response hand incongruency with these effects (2). To this aim, we merged responses to left and right hands in each rotation for human and clock stimuli in congruent and incongruent conditions, and the percentage of clockwise responses was computed. Since biomechanics constrains human hand movements in opposite directions for right and left hands, we flipped the left hand responses (i.e., when a clockwise response was prompted, we scored it as counterclockwise and vice versa). Otherwise, opposite responses to right and left human hands following biomechanical constraints would have artificially led to an average response of 50% (i.e., chance level). Moreover, we averaged the responses independently of the ISI since we did not expect this factor to impact the results. All our rotations implied the same range of movement (180°) regardless of the direction, and we aimed to verify whether body rotations were constrained to a different extent compared to non-corporeal objects like a clock. Thus, we entered these averaged values in a 2 × 2 × 8 ANOVA with the following within-subject factors: STIMULUS (Clock, Human), CONGRUENCY (Congruent, Incongruent), and ROTATION (0°−180°, 45°−225°, 90°−270°, 135°−315°, 180°−0°, 225°−45°, 270°−90°, 315°−135°).
As regards the first aim of Experiment 1, the presence of a significant main effect of STIMULUS should test our hypothesis of a significant difference between human and clock hands. Specifically, if the hands are subject to stronger bias induced by biomechanical constraints (i.e., leading to some rotations being constrained clockwise and others counterclockwise), this should result in an average clockwise response of approximately 50%. Conversely, if the clock hands exhibit a bias, it should favor the more familiar clockwise rotation. Thus, if our hypothesis is verified, a greater percentage of clockwise responses should be measured for clock hand stimuli. Moreover, in line with this reasoning, a significant interaction between STIMULUS and ROTATION could be hypothesized, with differences between hand and clock hand responses occurring specifically in those rotations where the human hand can move only in the counterclockwise direction.
As regards the second aim of Experiment 1, the possible interference effect of observing a human hand stimulus and responding with the other hand should manifest as a three-way STIMULUS × CONGRUENCY × ROTATION interaction. Indeed, if incongruency reduces the effect of biomechanical constraints, we should expect less clockwise (closer to 50%) responses in rotations where the constraint suggests a clockwise direction (0°−180°, 45°−225°) and more clockwise (closer to 50%) responses in rotations where the constraint suggests a counterclockwise direction (225°−45°, 180°−0°) only for human hand stimuli.
Experiment 2
Experiment 2 addressed whether visual priming affected the perception of constrained and unconstrained human and clock hand apparent rotations to the same extent. The percentage of clockwise responses was entered in a 2 × 2 × 2 ANOVA, with STIMULUS (Clock, Human), ROTATION (45°−225°, 135°−315°), and PRIME (Clockwise, Counterclockwise) as within-subject factors. A main effect of PRIME, with a greater percentage of clockwise responses in clockwise-primed conditions, should indicate that our priming task was effective irrespective of the stimulus kind (human, clock). Moreover, if the perception of human and clock apparent rotations was differently modulated by visual priming in constrained (45°−225°) or unconstrained (135°−315°) rotations, we predicted to observe a significant three-way STIMULUS × ROTATION × PRIME interaction.
Control experiment
Following Experiment 2, the very same paradigm was presented again to the same experimental subjects, but, in this case, participants were informed about the existence of a prime stimulus preceding the starting point of each rotation. They were asked to perform a forced choice task by reporting whether a rightward or leftward image was presented as a prime, and this happened even when they stated that they could not detect the orientation of the prime image. The accuracy of the participants’ responses was assessed using a one-sample Wilcoxon Sign Rank test against 50%.
Data availability
The datasets generated and/or analyzed during the current study are available in the Mendeley data repository at the following link: https://data.mendeley.com/preview/zk6f3y2358?a=d0b132a7-ca7b-4070-8567-37133810a517.
References
Blake, R. & Shiffrar, M. Perception of human motion. Annu. Rev. Psychol. 58, 47–73 (2007).
Chatterjee, S. H., Freyd, J. J. & Shiffrar, M. Configural processing in the perception of apparent biological motion. J. Exp. Psychol. Hum. Percept. Perform. 22, 916–929 (1996).
Kourtzi, Z. & Shiffrar, M. Dynamic representations of human body movement. Perception 28, 49–62 (1999).
Orgs, G., Bestmann, S., Schuur, F. & Haggard, P. From body form to biological motion: The apparent velocity of human movement biases subjective time. Psychol. Sci. 22, 712–717 (2011).
Shiffrar, M. & Freyd, J. J. Apparent motion of the human body. Psychol. Sci. 1, 257–264 (1990).
Shiffrar, M. & Freyd, J. J. Timing and apparent motion path choice with human body photographs. Psychol. Sci. 4, 379–384 (1993).
Verfaillie, K. & Daems, A. Representing and anticipating human actions in vision. Vis. cogn. 9, 217–232 (2002).
Kourtzi, Z. & Kanwisher, N. Activation in human MT/MST by static images with implied motion. J. Cogn. Neurosci. 12, 48–55 (2000).
Orgs, G., Kirsch, L. & Haggard, P. Time perception during apparent biological motion reflects subjective speed of movement, not objective rate of visual stimulation. Exp. Brain Res. 227, 223–229 (2013).
Vandenberghe, A. & Vannuscorps, G. Predictive extrapolation of observed body movements is tuned by knowledge of the body biomechanics. J. Exp. Psychol. Hum. Percept. Perform. 49, 188–196 (2022).
Wilson, A. D., Snapp-Childs, W. & Bingham, G. P. Perceptual learning immediately yields new stable motor coordination. J. Exp. Psychol. Hum. Percept. Perform. 36, 1508–1514 (2010).
Vannuscorps, G. & Caramazza, A. The origin of the biomechanical bias in apparent body movement perception. Neuropsychologia 89, 281–286 (2016).
Chen, J., Hu, Q. & Shao, Y. A letter on a hand: Rotating together or separately?. Q. J. Exp. Psychol. 72, 472–480 (2019).
Ionta, S., Fourkas, A. D., Fiorio, M. & Aglioti, S. M. The influence of hands posture on mental rotation of hands and feet. Exp. Brain Res. 183, 1–7 (2007).
Petit, L. S., Pegna, A. J., Mayer, E. & Hauert, C. A. Representation of anatomical constraints in motor imagery: Mental rotation of a body segment. Brain Cogn. 51, 95–101 (2003).
Seegelke, C. & Hughes, C. M. L. The influence of action possibility and end-state comfort on motor imagery of manual action sequences. Brain Cogn. 101, 12–16 (2015).
Thayer, Z. C. & Johnson, B. W. Cerebral processes during visuo-motor imagery of hands. Psychophysiology 43, 401–412 (2006).
Orgs, G. et al. Constructing visual perception of body movement with the motor cortex. Cereb. Cortex 26, 440–449 (2016).
Han, Q., Gandolfo, M. & Peelen, M. Prior knowledge biases the perception of body postures. J. Vis. 23, 4806 (2023).
Stevens, J. A., Fonlupt, P., Shiffrar, M. & Decety, J. New aspects of motion perception: Selective neural encoding of apparent human movements. NeuroReport 11, 109–115 (2000).
Vannuscorps, G., Pillon, A. & Andres, M. Effect of biomechanical constraints in the hand laterality judgment task: Where does it come from?. Front. Hum. Neurosci. 6, 1–20 (2012).
Berlucchi, G. & Aglioti, S. M. The body in the brain revisited. Exp. Brain Res. 200, 25–35 (2010).
de Vignemont, F. Body schema and body image-Pros and cons. Neuropsychologia 48, 669–680 (2010).
Gallese, V. & Sinigaglia, C. The bodily self as power for action. Neuropsychologia 48, 746–755 (2010).
Reed, C. L. & Vinson, N. G. Conceptual effects on representational momentum. J. Exp. Psychol. Hum. Percept. Perform. 22, 839–850 (1996).
Vinson, N. G. & Reed, C. L. Sources of object-specific effects in representational momentum. Vis. cogn. 9, 41–65 (2002).
Martín, A., Chambeaud, J. G. & Barraza, J. F. The effect of object familiarity on the perception of motion. J. Exp. Psychol. Hum. Percept. Perform. 41, 283–288 (2015).
Freyd, J. J. & Finke, R. A. Representational momentum. J. Exp. Psychol. Learn. Mem. Cogn. 10, 126–132 (1984).
Jones, H. G. et al. The effect of handedness on mental rotation of hands: A systematic review and meta-analysis. Psychol. Res. 85, 2829–2881 (2021).
Currò, T., Candidi, M. & Calvo-Merino, B. Inhibitory mechanisms are affected by stimulus-response congruency. Curr. Res. Behav. Sci. 4, 100108 (2023).
Anstis, S. & Ramachandran, V. S. Visual inertia in apparent motion. Vision Res. 27, 755–764 (1987).
Blake, R., Ahlström, U. & Alais, D. Perceptual priming by invisible motion. Psychol. Sci. 10, 145–150 (1999).
Cattaneo, L. et al. One’s motor performance predictably modulates the understanding of others’ actions through adaptation of premotor visuo-motor neurons. Soc. Cogn. Affect. Neurosci. 6, 301–310 (2011).
Davidenko, N., Heller, N. H., Cheong, Y. & Smith, J. Persistent illusory apparent motion in sequences of uncorrelated random dots. J. Vis. 17, 1–17 (2017).
Jiang, Y., Pantle, A. J. & Mark, L. S. Visual inertia of rotating 3-D objects. Percept. Psychophys. 60, 275–286 (1998).
Pantle, A. J., Gallogly, D. P. & Piehler, O. C. Direction biasing by brief apparent motion stimuli. Vision Res. 40, 1979–1991 (2000).
Verstraten, F. A. J., Cavanagh, P. & Labianca, A. T. Limits of attentive tracking reveal temporal properties of attention. Vision Res. 40, 3651–3664 (2000).
Dehaene, S. et al. Imging unconscious semantic priming. Nature 395, 597–600 (1998).
Sumner, P. Negative and positive masked-priming - Implications for motor inhibition. Adv. Cogn. Psychol. 3, 317–326 (2007).
Heller, N. H. & Davidenko, N. Dissociating higher and lower order visual motion systems by priming illusory apparent motion. Perception 47, 30–43 (2018).
Davidenko, N., Heller, N. H., Schooley, M. J. & McDougall, S. G. Visual priming of two-step motion sequences. J. Vis. 22, 1–21 (2022).
Giese, M. A. & Poggio, T. Cognitive neuroscience: Neural mechanisms for the recognition of biological movements. Nat. Rev. Neurosci. 4, 179–192 (2003).
Vannuscorps, G. & Caramazza, A. Typical action perception and interpretation without motor simulation. Proc. Natl. Acad. Sci. U. S. A. 113, 86–91 (2016).
Schütz-Bosbach, S. & Prinz, W. Perceptual resonance: Action-induced modulation of perception. Trends Cogn. Sci. 11, 349–355 (2007).
Rizzolatti, G. & Sinigaglia, C. The functional role of the parieto-frontal mirror circuit: Interpretations and misinterpretations. Nat. Rev. Neurosci. 11, 264–274 (2010).
Rizzolatti, G. & Sinigaglia, C. The mirror mechanism: A basic principle of brain function. Nat. Rev. Neurosci. 17, 757–765 (2016).
Anderson, S. F., Kelley, K. & Maxwell, S. E. Sample-size planning for more accurate statistical power: A method adjusting sample effect sizes for publication bias and uncertainty. Psychol. Sci. 28, 1547–1562 (2017).
Oldfield, R. The assessment and analysis of handedness: The Edinburgh inventory. Neuropsychologia 9, 97–113 (1971).
Tidoni, E., Scandola, M., Orvalho, V. & Candidi, M. Apparent biological motion in first and third person perspective. Iperception. 7, 1–6 (2016).
Acknowledgements
This article was supported by the Department of Philosophy ‘Piero Martinetti’ of the University of Milan with the Project “Departments of Excellence 2018-2022” awarded by the Italian Ministry of Education, University and Research (MIUR) (to CS), by the ERC-STG “MyFirstBody” (101078497 to FG), and by the PRIN 2017 project “The cognitive neuroscience of interpersonal coordination and cooperation: a motor approach in humans and non-human primates” (201794 KEER; to CS). However, the views and opinions expressed are those of the authors only and do not necessarily reflect those of the European Union or the European Research Council Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.
Author information
Authors and Affiliations
Contributions
Conceptualization, S.P., C.S., L.C., and F.G.; Methodology, A.R.S., M.R., and S.P.; Software, A.R.S., M.R., and S.P.; Formal analysis, A.R.S., and M.R.; Investigation, A.R.S., M.R.; Writing –Original Draft, C.S., A.R.S., M.R., and S.P.; Writing – Review and Editing, C.S., A.R.S., M.R., S.P., F.G. and L.C.; Visualization, A.R.S.; Supervision, S.P., C.S., L. C. and F.G.; Funding Acquisition C.S. and F.G.
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Parmigiani, S., Sebastiano, A.R., Romeo, M. et al. How body knowledge shapes motion perception. Sci Rep 15, 16878 (2025). https://doi.org/10.1038/s41598-025-00213-0
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s41598-025-00213-0
