Training variability has been brought forward as one possible moderator for wider scale transfer effects in cognitive training. However, little is known about which aspects of task variability are important for optimizing training outcomes. This study systematically examined the impact of variability in the different task components on outcome measures, here manipulating content (whether the task stimuli remained fixed or changed between blocks) and the deeper structural task configuration (task sequence: whether the task sequence was fixed or random). Short-term task switching training was implemented with one of four training variability conditions: fixed content\fixed structure; fixed content\ random structure; varied content\fixed structure and varied content\varied structure. The experiment consisted of a baseline block, seven training blocks (learning phase), followed by two transfer blocks, one with fixed and one with random task structure, respectively. In the learning phase, more rapid training gains were observed in the fixed content as compared to varied content. Interestingly, training with fixed content resulted in a trend for costs when transferred to a novel task switching context. In contrast, moderate transfer gains were noted in the varied content condition, manifested specifically on switch trials. These results suggest that task (content) variability is one of the means to improve positive transfer and avoid negative transfer. Additionally, and in agreement with the wide literature on training, this finding suggests that conditions that prevent training gains are in fact beneficial for learning generalization.

Content variability was previously suggested to promote stronger learning effects in cognitive training whereas less variability incurred transfer costs (Sabah et al. Psychological Research, 10.1007/s00426-018-1006-7, 2018). Here, we expanded these findings by additionally examining the role of learners’ control in short-term task-switching training by comparing voluntary task-switching to a yoked control forced task-switching condition. To this end, four training conditions were compared: (1) forced fixed content, (2) voluntary fixed content, (3) forced varied content, and (3) voluntary varied content. To further enhance task demands, bivalent stimuli were used during training. Participants completed baseline assessment commencing with task-switching and verbal fluency blocks, followed by seven training blocks and last by task-switching (near transfer) and verbal fluency (far transfer) blocks, respectively. For the baseline and transfer task-switching blocks, we used the exact same baseline and first transfer block from Sabah et al. (Psychological Research, 10.1007/s00426-018-1006-7, 2018), employing univalent stimuli and alternating-runs task sequence. Our results pointed again to the contribution of content variability to task-switching performance. No indications for far transfer were observed. Allowing for learners’ control was not found to produce additional transfer gains beyond content variability. A between-study comparison suggests that enhanced task demands, by means of bivalency, promoted higher transfer gains in the current study when compared to Sabah et al. (Psychological Research, 10.1007/s00426-018-1006-7, 2018). Taken together, the current results provide further evidence to the beneficial impact of variability on training outcomes. The lack of modulatory effect for learners’ control is discussed in relation to possible methodological limitations.

Internal working memory (WM) gating control policies have been suggested to constitute a critical component of task-sets that can be learned and transferred to very similar task contexts (Bhandari and Badre (Cognition, 172, 33–43, 2018). Here, we attempt to expand these findings, examining whether such control policies can be also trained and transferred to other untrained cognitive control tasks, namely to task switching and AX-CPT. To this end, a context-processing WM task was used for training, allowing to manipulate either input (i.e., top-down selective entry of information into WM) or output (i.e., bottom-up selective retrieval of WM) gating control policies by employing either a context-first (CF) or context-last (CL) task structure, respectively. In this task, two contextual cues were each associated with two different stimuli. In CF condition, each trial began with a contextual cue, determining which of the two subsequent stimuli is target relevant. In contrast, in the CL condition the contextual cue appeared last, preceded by a target and non-target stimulus successively. Participants completed a task switching baseline assessment, followed by one practice and six training blocks with the WM context-processing training task. After completing training, task-switching and AX-CPT transfer blocks were administrated, respectively. As hypothesized, compared to CL training condition, CF training led to improved task-switching performance. However, contrary to our predictions, training type did not influence AX-CPT performance. Taken together, the current results provide further evidence that internal control policies are (1) inherent element of task-sets, also in task switching and (2) independent of S-R mappings. However, these results need to be cautiously interpreted due to baseline differences in task-switching performance between the conditions (overall slower RTs in the CF condition). Importantly though, our results open a new venue for the realm of cognitive enhancement, pointing here for the first time to the potential of control policies training in promoting wider transfer effects.

In our daily life, we often encounter situations in which different features of several multidimensional objects must be perceived simultaneously. There are two types of environments of this kind: environments with multidimensional objects that have unique feature associations, and environments with multidimensional objects that have mixed feature associations. Recently, we (Goldfarb & Treisman, 2013) described the association effect, suggesting that the latter type causes behavioral perception difficulties. In the present study, we investigated this effect further by examining whether the effect is determined via a feedforward visual path or via a high-order task demand component. In order to test this question, in Experiment 1 a set of multidimensional objects were presented while we manipulated the letter case of a target feature, thus creating a visually different but semantically equivalent object, in terms of its identity. Similarly, in Experiment 2 artificial groups with different physical properties were created according to the task demands. The results indicated that the association effect is determined by the task demands, which create the group of reference. The importance of high-order task demand components in the association effect is further discussed, as well as the possible role of the neural synchrony of object files in explaining this effect.

The ability to learn abstract generalized structures of tasks is crucial for humans to adapt to changing environments and novel tasks. In a series of five experiments, we investigated this ability using a Rapid Instructed Task Learning paradigm (RITL) comprising short miniblocks, each involving two novel stimulus-response rules. Each miniblock included (a) instructions for the novel stimulus-response rules, (b) a NEXT phase involving a constant (familiar) intervening task (0-5 trials), (c) execution of the newly instructed rules (2 trials). The results show that including a NEXT phase (and hence, a prospective memory demand) led to relatively more robust abstract learning as indicated by increasingly faster responses with experiment progress. Multilevel modeling suggests that the prospective memory demand was just another aspect of the abstract task structure which has been learned.