Organization, Learning and Action
Independence between action components in catching revealed by visual uncertainty.
In interception the timing of the action to catch a ball at the arrival time is controlled through two components: RT (catch initiation) and movement time (MT). Uncertainty of the arrival time could affect either component or both. By showing different parts of the parabolic path of a ball I manipulated the degree of uncertainty of visual estimates of arrival time and studied how this manipulation affects RT and MT. Subjects wearing a data glove had to catch balls in a VR setting. The trajectories were shown in 3 viewing conditions (early, late and full path) and haptic feedback was given in half of the sessions. Regardless of the feedback, the mean RT and its SD were affected by visual uncertainty: more uncertainty led to start closing the hand earlier and with higher variability. When no haptic feedback was provided, there was a RT-MT coupling, in which uncertainty of RT predicted MT duration (slope about 1). However when haptic feedback was given, MT duration was independent of the initiation of the catch. This suggests that unfolding the catch is controlled by the expected sensory consequences, while a threshold based on different visual information triggers the initiation of the catch.
Temporal dynamics of candidate interpretations during object recognition
Katrien Torfs, Sven Panis, Annelies Bartlema and Johan Wagemans
We investigated the role of candidate interpretations when people recognize objects, focusing on the contribution of different spatial frequencies in their formation, and the time-windows in which they are most effective. We used a stimulus set consisting of 80 gray-scale pictures of objects, and filtered versions with either their low or high spatial frequencies (LSF and HSF, resp.). In a pre-study with long exposures, subjects (N=105) were asked to list all potential objects that could have given rise to the LSF pictures. In two priming experiments with short exposures, subjects had to classify the objects as manmade or natural (Experiment 1, N=20), or to identify them (Experiment 2, N=7). Survival analysis was used to study the temporal dynamics of candidate interpretations during object recognition. The results showed that the number of candidates generated by the LSFs of the target influences the occurrence of recognition responses, but only when response latency is long, and all candidates belong to the same category as the target. Our results highlight the importance of coarse shape information in the activation of candidate interpretations during object recognition and, more generally, the need to control for the moment of responding by using an appropriate statistical technique.
Adaptation to temporal delays requires dynamic feedback
Cristina de La Malla, Joan Lopez-Moliner and Eli Brenner
It is known that people can adapt to temporal delays. We here examine what kind of feedback promotes temporal adaptation. We asked participants to hit targets that were moving in different directions at different speeds, and compared performance across sessions in which (a) delayed visual feedback was provided throughout the movement of the hand, (b) participants saw a brief delayed representation of their hand as it crossed the target, and finally (c) they were shown the spatial error (knowledge of result) in their performance immediately after each movement. Participants could only correct their on-going movements in the first condition, so performance in the other two relied totally on adapting to the delay. The temporal delay was gradually increased to reach a total of 200ms. The timing of the hand movement with full feedback closely followed the imposed delay. This was also so when participants only saw the last part of the movement. Much less adaptation was found when the positions of the target and hand at the time of the hit were displayed statically after each movement. We conclude that feedback with spatiotemporal structure is necessary for temporal adaptation.
Objects affords kinematic properties of potential actions
Aurelien Richez, Yann Coello and Gérard Olivier
In this study, we analysed the effect of object affordances on the motor response to intrinsic (colour) property of a visual object. Previous studies have shown that the compatibility between a manual response to an (intrinsic or extrinsic) object property and the action afforded by the object led to a decrease of response time. In the present study, we assessed whether afforded action can also influence the execution of a motor response when processing intrinsic (non-motor) property of a visual target. 32 participants performed a colour discrimination task of black and white chess pieces presented at different distances on a chessboard. Participants had to grasp as quickly as possible either a proximal or a distal switch in response to the colour of the chess piece. We manipulated the compatibility between the action afforded by the target-object and the response required by the perceptual task. Results showed a motor compatibility effect on reaction time, as previously shown, but also on movement time, suggesting that objects afford not only potential actions but also their dynamical properties.
Can people change their choice of grasping points during the grasping movement?
Dimitris Voudouris, Jeroen B.J. Smeets and Eli Brenner
To grasp an object, one must select suitable positions on its surface for placing the digits and move the digits to those positions. We know that digits'movements are adjusted if the grasping points move, but do people always stick to their original choice of grasping points or do they readjust them if doing so is beneficial? Subjects grasped a ball or cube that sometimes rotated briefly by a few degrees immediately after movement onset. When the cube rotated, subjects rotated their grip to end with the digits at about the same points on the surface. When the ball rotated, subjects did not change their movement but simply grasped the ball at different points on its surface. Does this mean that people can reconsider the grasping points during the movement, or are the grasping points defined in relation to the overall shape (that does not change when the ball rotates) rather than to the surface texture? We rotated the cube after carefully placing it such that it could be grasped with one of two orthogonal grip orientations. In some trials, subjects switched grip orientations when the cube was rotated. Thus people can strategically alter the grasping points during the movement.
Visual and visuomotor perceptual learning: two distinct mechanisms
Lukasz Grzeczkowski, Fred W. Mast and Michael H. Herzog
We investigated if visual and visuomotor perceptual learning rely on the same underlying mechanisms. Observers trained with line bisection stimuli in which a center line was either closer to a right or left outer line. In the visuomotor learning condition, observers adjusted the center line with the computer mouse to the smallest offset they could perceive. In the visual learning condition, observers indicated in a binary task if the center line was closer to left or right outer line. Learning occurred in both conditions. Interestingly, improvements in either condition did not transfer to the other condition as shown by pre- and post-measurements (i.e. visuomotor learning did not improve visual performance and visual learning did not improve visuomotor performance). Our results suggest that visual and visuomotor perceptual learning rely on different mechanisms.
Motor action reduces the temporal asynchrony between two visual but not visuo-tactile changes.
Xavier Corveleyn, Joan Lopez-Moliner and Yann Coello
Perceiving a visual object requires binding of its various physical attributes. However, temporal order judgments (TOJ) have revealed temporal asynchronies in the perception of position and colour changes of a visual target. This study analysed the effect of a motor action on these temporal asynchronies. Results showed that in the perceptual condition, colour change must precede (-37.9ms) position change in order to participant perceive a synchronous change. This physical asynchrony nearly vanished when the same changes took place near the end of a manual reaching action executed towards the visual target (-3.3ms). This reduction of asynchrony was however not observed when participants performed TOJ of visual attributes change in the presence of concomitant tactile information (-36 ms) but with no action. The perceptual relative timing between visual changes was also unaffected when the timing was obtained by comparing each visual change to tactile information resulting from motor action (finger-target contact, -33.5ms) or concomitant tactile information (-27.8ms). These results suggest that predictive signals associated to motor actions contribute to reduce the differential delays when processing two visual attributes of a single object, but this effect does not propagate to cross-modal timing.
Modality differences in rhythm perception of human movement
Mauro Murgia and Penny McCullagh
Humans generally discriminate auditory rhythmical sequences better than visual ones. Previous researches focused mainly on laboratory tasks. We extended this domain to a complex motor task. In two experiments, we tested 52 subjects, expert tap dancers and novice people, who were exposed to movies and soundtracks of paddle sequences performed by a tap dance instructor. The fifty percent of the sequences followed an even rhythm, the remaining fifty percent presented a single step with a rhythmical error included in a range between twelve and twenty percent, compared to the even performance (uneven trials). The task of the subjects was to discriminate the rhythm of performances by expressing a dichotomous response: 'even' or 'uneven'. The results evidenced responses superior to the chance level in all the auditory conditions and in the visual condition with errors superior to sixteen percent. The accuracy was constantly better in the auditory conditions, rather than in the visual conditions. Expertise differences emerged only in the auditory condition. These outcomes are consistent with the literature and confirm the elevated sensitivity of the acoustic channel to rhythmical stimuli. These findings suggest that the temporal information of action could be easily inferred by athletes through auditory models.
How perceptual learning shapes perception
Céline Cappe, Christine Mohr and Michael Herzog
Lifelong learning makes experts. Here, we asked the question how every day experience shapes vision through perceptual learning. We tested performance in visual acuity, vernier discrimination, visual backward masking, Gabor contrast detection, and bisection discrimination tasks. When lifelong visual learning leads to generalization on basic skills, we expect strong correlations between tasks. When perceptual learning is specific, we expect no or very little correlations. The latter is indeed what we found in 36 healthy student observers (mean age = 21.1 years). Except for Gabor contrast detection and visual acuity, all other pair wise correlations were non-significant. These results cannot be explained by intra-observer variability because performance within one task was highly reproducible. In summary, our study suggests that everyday experience shapes perception in a very specific manner.
Perceptual learning can be improved by transcranial random noise stimulation
Anna Fertonani, Cornelia Pirulli and Carlo Miniussi
Perceptual learning is considered a manifestation of neural plasticity in the human brain. We investigated brain plasticity mechanisms in a visual perceptual learning task using non-invasive transcranial electrical stimulation (tES, i.e., direct current stimulation tDCS and random noise stimulation tRNS) applied on the visual areas. Our hypothesis is that different types of tES would have varying actions on the nervous system, which would result in different efficacies of neural plasticity modulation. 107 healthy volunteers participated in the experiment, assigned to the groups: high-frequency tRNS (hf-tRNS, 100-640 Hz), low-frequency tRNS (lf-tRNS, 0.1-100 Hz), anodal-tDCS (a-tDCS), cathodal-tDCS (c-tDCS) and sham. Furthermore, a control group was stimulated on the vertex (Cz). The analysis showed a learning effect during task execution that was differentially modulated according to the stimulation conditions. Post-hoc comparisons revealed that hf-tRNS significantly improved performance accuracy compared with a-tDCS, c-tDCS, sham and Cz stimulations. Our results confirmed the efficacy of hf-tRNS over the visual cortex in improving behavioral performance and showed its superiority in comparison to others tES. The repeated subthreshold stimulation of tRNS may prevent homeostasis of the system and potentiate task-related neural activity. This result highlights the potential of tRNS and advances our knowledge on neuroplasticity induction approaches.
Which is the best timing for neuroplasticity induction in a perceptual learning task?
Cornelia Pirulli, Anna Fertonani and Carlo Miniussi
Transcranial electrical stimulation (tES) induces changes in the brain activity, leading to alterations in the behavioral tasks performance. Previous studies suggest the importance of timing in the application of the stimulation. We hypothesize that in the visual system the effects of tES (i.e., direct current stimulation - tDCS and random noise stimulation - tRNS) are dependent on the state of cerebral activation. Therefore the same type of stimulation, applied in different moments, may have different effects. We applied tRNS and tDCS before (offline) or during (online) a visual perceptual learning task. 106 participants were divided in seven groups: anodal-tDCS (a-tDCS) offline and online, cathodal-tDCS (c-tDCS) offline and online, tRNS offline and online and sham stimulation. Our results confirm that exists an ideal timing of application, depending on the type of stimulation. We observed an improvement of the performance when a-tDCS was applied before the task, whereas with tRNS we had a great improvement in the performance only during the task. Surprisingly an analogous improvement was present after offline c-tDCS. These results are important for the designing of rehabilitation protocols, highlighting which stimulation is better to choose in relation to its timing of application.
Illusory Contours Obtained by Means of Cast Shadows
Among the procedures for obtaining illusory contours there is an original one obtained by depicting appropriately arranged black cast shadows. Various examples have been created by the author and are shown in the article 'From Visual Art to Experimental Psychology and Viceversa' [Bonaiuto, 2011, in: Affective Processes, Cognition and Action, V. Biasi, Rome, Teseo]. The most refined example depicts athletes in a hurdle race (see the figure). The hurdles have been created by means of illusory contours through cast shadows. A black shadow of proportionate length extends from the propelling foot of each athlete and then appears to bend and to outline a barrier at the other end. These barriers are aligned in parallel to form a row of coordinated elements. The shadows appear to systematically bend and to outline the object which appears to be the cause of this bending, that is, the barrier of hurdles. One of the runners appears to be suspended in mid air while jumping over a hurdle, while the others have different positions corresponding to the various moments of their run-up. The drawing also has an autobiographical reference since the author was a regional champion in Italy in the 400-metre hurdles race.
Anomalous structure from motion from figure-ground
Vicky Froyen, Jacob Feldman and Manish Singh
We present a novel phenomenon involving an interaction between motion contrast and figure-ground. Our displays contain alternating light and dark vertical strips in which random dot textures moved horizontally, in opposite directions in alternating strips. This motion is consistent with a percept of all light strips in front, with the dark strips completing amodally into a single large moving surface in the back; or vice versa. Surprisingly, the strips that are perceived as figural are also perceived as 3D volumes rotating in depth (like rotating cylinders)---despite the fact that dot motion is not consistent with 3D rotation. (All dots within a strip have the same constant velocity; and any strip shown in isolation is perceived as flat.) We found we could easily manipulate which set of strips is perceived as rotating volumes simply by varying known geometric cues to figure and ground. In experiments, we varied convexity, parallelism, symmetry, and proximity. Subjects indicated which colored strips they perceived as rotating. The results nicely tracked these classic figure-ground cues. In our data, convexity was a substantially stronger cue to figure-ground than either symmetry or parallelism. Our results reveal an interesting interaction between motion contrast and figure-ground, and provide an effective tool for measuring figure-ground perception.
Rectangle transformations and size bisection
Walter Gerbino and Giacomo Bulgarelli
The domain of rectangles can be described with reference to two transformations: the constant ratio transformation (CRT), in which the aspect ratio is invariant over size change while adjacent sides grow at different rates; and the medial axis transformation (MAT), in which the aspect ratio covaries with size while adjacent sides grow at a constant rate. The square is singular inasmuch as the products of CRT and MAT transformations coincide. To obtain evidence of the relative importance of such transformations for shape perception we used data from the bisection of size intervals along CRT vs. MAT rectangle continua. Participants were required to adjust a target rectangle until its size appeared intermediate between those of two extreme rectangles (one about three times larger than the other). Surprisingly, bisection errors for CRT continua were inconsistent; while those for MAT continua displayed a regular increase as a function of rectangle elongation (i.e., of the amount of discrepancy from the square). As regards the phenomenal shape of rectangles, including the square, the invariant growth of sides seems more important than their invariant ratio.
Perception of shape in apparent motion displays
Aleksandra Zharikova, Sergei Gepshtein and Cees van Leeuwen
We studied how the perception of shape depends on spatial and temporal grouping of elements that make up a closed contour. Observers viewed dynamic displays in which several dots were moving, each along a linear or slightly curved trajectory. In one of the frames ('keyframe') the dots formed the target: an oriented shape tilted clockwise or counterclockwise from the vertical. Observers reported the direction of tilt and rated their confidence on a six-point scale. We varied two parameters. First is shape coherence determined by the distance between actual dot locations and the target shape in the keyframe. As coherence decreased the shape became less salient. Second is distance dM between successive dots within their trajectories. The threshold of shape coherence (at 75% correct discrimination) was a function of dM, indicating that salient apparent motion could disrupt perception of shape. We varied the temporal rate of frames in these displays and also measured strength of apparent motion in displays that contained no shape. This way we isolated contributions of two kinds of grouping: spatiotemporal grouping between frames and spatial grouping within keyframe, and thus measured observers'sensitivity to shape separately from their sensitivity to motion.
Perception of angles between surfaces is biased towards right angles
Torsten Betz and Marianne Maertens
If 2D perspective projections of two rectangular surfaces joined at various angles, like folded cards, are viewed monocularly, they appear to always enclose a right angle. In addition, surfaces enclosing an acute angle seem to look taller than wide, while surfaces enclosing an obtuse angle look wider than tall. We reasoned that if the perceived angle is used as a visual cue to infer the true size of the projected surface, and the angle is perceived incorrectly, then angle variations should be accompanied by systematic misjudgement of perceived surface size, as predicted by a bias to perceive the enclosed angle as perpendicular. We tested our hypothesis by asking observers to adjust the width of a comparison surface to match the aspect ratio of previously shown test surfaces. Observers monocularly viewed projections of two adjoining surfaces of five different widths and three enclosed angles on a computer-screen. Observers judged object width correctly for the right angle. Object width was systematically underestimated for surfaces enclosing 60° angles and overestimated for surfaces enclosing 120° angles. We conclude that with monocular cues only, observers are biased towards perceiving angles between surfaces as perpendicular, and that this misperception affects how they perceive other stimulus properties.
Is there a common underlying factor in global vs. local biases in perceptual organization?
Lee De-Wit, Sander Van de Cruys, Lore Verhoogen and Johan Wagemans
It is becoming evident that people's perceptual abilities vary at an individual level in ways that can be meaningfully related to the neural architecture of the human visual system. Indeed, in the context of research on Autism, it has been argued that there may exist a continuum of variability with regards to whether people focus on local details or the global whole. This general approach (and Weak Central Coherence in particular), however, is often only tested in the context of one stimulus (e.g., embedded figures). This selective use of only one or two visual stimuli in a given study makes it hard to evaluate the general claim that there is such a thing as an underlying bias with regards to perceptual organization. To address this, the current work tests the bias towards more local or global interpretations with 10 different cases of perceptual organization (incl. Kanizsa figures, amodal completion, Mooney figures, Stuart Anstis's balls and grouping in Multiple Object Tracking). These different cases of perceptual organization were tested on 190 University of Leuven undergraduates and will be compared with a common measure of autistic characteristics in the general population (the Autism Quotient).
Multisensory interaction in prism adaptation
Manfred Fahle, Mona Bornschlegl and Gordon Redding
Prism adaptation is used to study the plasticity of eye-hand coordination to misalignment of the visual and proprioceptive spatial maps. Misalignment can be resolved by adaptive change in spatial maps of either the eyes or hand or both. In this procedure pacing pointing movements with a rhythmic auditory signal is usually employed to control movement speed, but the role of the auditory signal itself in producing adaptation has not been examined. The present experiment addressed this issue by testing three conditions: (1) exposure pointing was self-paced without an auditory signal, (2) exposure pointing was paced by an auditory signal without synchronization, and (3) exposure pointing was synchronized with the auditory signal. The first condition produced primarily proprioceptive adaptation. The second condition also produced primarily proprioceptive adaptation, but visual adaptation was also present. The third condition produced primarily visual adaptation. Results are discussed in terms of a possible role for the auditory signal: movement synchronization with a rhythmic auditory signal may enable multisensory integration, including auditory spatial information that selects the more reliable proprioceptive signal for movement control. Consequently, detection of the misalignment is localized and realignment occurs in the visual system.
Assessing near vision function: The Italian version of the Radner Reading Chart
Alessandro Fossetti, Antonio Calossi, Laura Boccardo and Wolfgang Radner
Purpose : As visual acuity tests are poor predictors of the real-world functions, performance-based tests, e.g., reading speed measurements, can be used to better quantify near visual function. The original German-language Radner Reading Chart emphasizes the principle of highly standardized sentences, i.e. highly comparable in number and length of words, as well as in difficulty and construction. The aim of this study was to create an Italian version of the Radner Reading Chart. Methods: 41 'sentence optotypes' with comparable structure and lexical and grammatical difficulty were created in Italian language, following the procedure defined by Radner. Sentences were performed in 211 normal, non-presbyopic, native Italian-speaking persons. The most equally matched sentence in terms of reading speed and number of reading errors were selected, to develop 28 short Italian sentences. The reading speed results obtained with the 28 selected short sentences were also compared with the average of two long 4th-grade paragraph (97 and 90 words) under the same conditions. Results: The overall mean reading speed of the tested persons was 189 +/- 26 w/min. The 28 sentences more similar as reading time were selected, achieving a coefficient of variation (the relative SD) of 2.2%. The reliability analyses yielded an overall Cronbach's alpha coefficient of 0.98. The correlation between the short sentences and the long paragraph was high (r= 0.85, p P<0.0001) Conclusions: The Italian Radner Reading Chart is precise (high consistency) and practical (short sentences) and therefore useful for research and clinical practice to simultaneously measure near reading acuity and reading speed.