Posters: Action I

Planning an action can be construed as the temporal binding of features of perceptual action effects. Previous research demonstrated such binding for task-relevant, body-related effect features. However, there is also evidence that binding does not encompass features of a task-irrelevant, body-external effect. In three experiments, we investigated whether it is task-relevance or body-relatedness that determines whether feature binding occurs. Participants prepared a certain action A, but before executing it, initiated an intermediate action B. Each action relied on a feature of a body-related effect (index vs. middle finger movement) and a feature of a body-external effect (cursor movement towards vs. away from a reference object). In Experiment 1 (n=34), both effects were equally task-relevant. Performance in action B suffered from partial feature overlap with action A, compared to full feature repetition/alternation, which indicates binding of both types of features in planning action A. Importantly, this partial overlap cost disappeared when both effects were available but only the body-related effect was task-relevant (Experiment 2, n=34). Moreover, when only the body-external effect of action A was known in advance (Experiment 3, n=36), performance in action B was better if it aimed at the same, rather than a different, distal effect. Consequently, task-relevance determines whether binding of body-related and body-external effect features takes place, while a lack of binding does not hinder the pre-activation of features of relevant body-external effects prior to action execution.

Binding theories postulate an automatic integration of stimulus and response features into temporary episodic traces (or event files). A repetition of any of the features from the previous episodic trace results in the retrieval of the entire trace. Along with relevant stimuli, even irrelevant stimuli or features are integrated into these episodic traces, and repeating irrelevant stimuli can also retrieve the previous episodic trace including the previous response – distractor-response binding. This binding is generally thought to be largely automatic in nature, occurring without any great requirement of attentional resources. However, some studies have shown that specific types of attention, e.g. spatial attention or feature-based attention can influence binding effects. In the present study, the role of central attention on distractor-response bindings was examined. To this end, participants carried out a primary letter identification task and a secondary updating task in parallel. Binding effects were tested under conditions of high cognitive load, low cognitive load and a control condition with no cognitive load. The results indicate smaller binding effects under conditions of high working memory load compared to a control condition, thus suggesting that although such bindings are automatic, they are not completely independent of central attentional resources.

During visual imagery, mental images are driven by internal signals in the absence of external inputs. Here we asked how categorical information is modulated by task demands. To this aim, we measured participants’ fMRI signal while they imagined static and dynamic animals and tools. We found that imagery and perception of tools in both tasks activate similar brain regions, involving superior parietal lobule (SPL) and lateral occipital cortex (LOC), but only ~23% of voxels overlapped between the two processes. Representational similarity analysis (RSA) furthermore demonstrated that the LOC captures information about object categories, irrespective of the task (static, dynamic). Additionally, RSA showed that information about imagined static objects is carried in the middle temporal gyrus, whereas information about imagined dynamic objects is spread across several brain regions. To examine the underlying functional connectivity, we used psychophysiological interaction (PPI) analysis. We found that the functional connectivity between animal-selective seeds and regions in the right ventral pathway was modulated by the difference between perception of animals and tools. Moreover, we found a decreased functional coupling between LOC and category-selective regions during visual imagery, and an increased coupling between SPL and the frontal cortex during static visual imagery. Our results highlight the differences between perception and visual imagery. Moreover, our results enhance our understanding of the neural basis of visual imagery, showing that imagined stimulus categories can be distinguished in the LOC based on patterns of activation, and that imagining dynamic objects engages a wider network than imagery of static objects.

Capacity limited cognitive functions seem to be particularly sensitive to the detrimental effects of mental fatigue induced by increasing Time-on-Task (ToT). Previous studies have also suggested that movement behavior, especially the preparatory phase, is costly in term of cognitive capacity. Yet effects of ToT specific to the different phases of movements have received little attention. Therefore, in two experiments, we assessed the effect of ToT on a visually guided pointing task. In both experiments, participants (n = 23 and 23) were instructed to point to targets by moving the cursor from the center to the peripheral target. In experiment 1, targets appeared at one of the four positions. In experiment 2, there were 16 target positions enhancing the uncertainty about movement direction. The first three blocks of trials lasted 15 minutes without rest. Participants then had a 2-min break followed by an additional block. Data of movement preparation time, movement execution, and subjective fatigue were recorded. Movement execution was measured as movement time, movement error, peak velocity, path length-task axis length ratio etc. Gaze position recording was also used to control fixation. In both experiments, the most robust finding was that movement preparation became slower with increasing ToT. In contrast, movement execution was associated with decreasing speed-accuracy trade-off: fatigued participants made faster but more erroneous movements. To conclude, the results suggest that enhanced level of mental fatigue is manifested in a slow preparatory phase followed by a faster but often more erroneous movement execution.

Research on memory for intentions and goals suggests that not only cognitive representations of intended actions but also semantic concepts that are associated with them exhibit heightened memory activation prior to completing an intention or reaching a goal. This has been observed, for instance, in faster lexical decisions for intention-related compared to unrelated memory contents. After intention completion, however, this so-called intention-superiority effect is presumed to dissipate or even reverse into lower activation of intention-related compared to unrelated memory contents (i.e., intention inhibition). Aiming to replicate these effects, in the present study, we tested the activation status of verbal intention representations across five experiments that were designed to detect both high intention activation before completion (i.e., intention superiority effects) and strong intention inhibition after completion. To our surprise, the present experiments showed that activation levels of intention-related and unrelated memory contents did not differ reliably prior to or after intention completion. This was also mirrored by Bayesian analyses and a meta-analytic integration of our findings, suggesting that these effects might be rather fragile and require further research to identify variables that modulate their occurrence.

Actions and their effects are perceived temporally shifted toward each other compared to the same events in isolation. While this phenomenon of temporal binding was initially discussed as a perceptual bias that is unique for intentional actions, similar observations have now also been made in other event chains. What brings about this effect, however, cannot yet be clearly pinpointed. In this study, we consider the temporal binding paradigm with respect to multisensory cue integration by manipulating temporal certainty of actions and their effects. We discuss the results in light of the “principle of inverse effectiveness” in multisensory cue integration which suggests that the perceived timing of less certain events is pulled strongly towards related and temporally certain events while temporally certain events are relatively unaffected by uncertain events.

The constant-foreperiod effect refers to an increase of reaction time (RT) with increasing foreperiod (FP) length. Proceeding from a pioneering study (Holender & Bertelson, 1975), we examined whether or not sequential action biases in choice-RT tasks remain stable over time (during the FP interval). In three experiments, we examined performance as a function of constant-FP length (1000 vs. 5000 ms) and the sequential effects of event repetitions versus alternations. As a result, sequential action biases occurred predominantly in short-FP trials and decreased in long-FP trials. Crucially, this interactive effect of FP length and action bias on performance was completely abolished when the intertrial interval was further increased. These results challenge the popular belief that event-specific contributions to performance are a stable part of the mental representation that guides temporal expectations in FP situations. Rather, they indicate that transient activation merely superimpose on performance effects.