“Capture and Control: Working Memory Modulates Attentional Capture by Reward-Related Stimuli”

“Capture and Control: Working Memory Modulates Attentional Capture by Reward-Related Stimuli”

The field of neurocognitive processes is one that elicits a lot of interest among researchers, especially in an attempt to understand common conditions associated with the reward system influence. In this paper, the difference between Value-driven attentional capture (VDAC) and saliency or contingency were of specific interest to the researchers Watson, Pearson, Chow, Theeuwes, Wiers, Most and Le Pelley (2019).  Their working hypothesis was that minimum interference by goal-irrelevant distractors is achievable through selective visual attention provided by goal-directed working memory.  The researchers went ahead to do experiments to test if resource-dependent control processes impact the presence of a reward on the oculomotor capture as much as they did on physically conspicuous objects. Their guiding question was whether a reward’s influence on the capture is an automatic process or the it can be reduced by cognitive control processes when the impact of the capture is not as per one’s current goals. With recent studies having demonstrated the association between reward and stimuli in attention capture, the study question is of importance as it helps provide more information on the behaviour of attentional bias as is the case in drug-related stimuli for addicts.

For this research, Watson, Pearson, Chow et al. (2019) performed two experiments. In the first experiment, the oculomotor capture, measured through the saccade made using a Tobii TX300 eye tracker was the dependent variable and the singleton distractors acted as the independent variables. The research involved 30 participants, each paid AUD 25 and a bonus dependent on their performance (the reward). The tests were administred individually with the eye tracker positioned on a 23-inch monitor, and the participant’s head placed 60 cm away from the screen in a chin rest position.

Each individual was to stare at a fixated central cross, which after 300ms changed colour to yellow to show the commencement of the visual search task.  Six shapes- a diamond and five circles placed evenly around an imaginary ring at an angle of 10.1 degrees were involved. In some trials, the circles were blue or orange, and the rest of the circles were grey. The coloured ones were the singleton distractors. Blue represented the top reward colour and, orange was the low, in half the group and vice versa. The participants would quickly look at the diamond, and the response captured after a gaze time of 100ms of gaze time towards the target centre. If the trial had a high reward singleton and rapid response was recorded, the participant earned 500 points, but if it was a low reward singleton or none, a fast response earned him/her 10 points. Gaze towards the distractor though would be an omission trial hence no reward.

For the memory task, in the high load tests, the trial begun by the participants were shown randomly arranged digits of 1 to 5 for about 0.1s then given a 0.05s blank interval before the search task was administered performed. As soon as giving the search task feedback, they were supposed to key in the number that appeared in one of the randomly chosen locations. A delay of 5s resulted in an error tune, also if the entry done was wrong. In a low load test, a similar test was done, but only one digit was shown out of the five at the beginning of the trial.  Before beginning the trials, the participants were explained to how the study would be done, and they were given a demonstration chance to ensure they had understood. Also, the information on conversion rates for their rewards was not given forth but they were informed that they would be given a bonus of between AUD 8 AND 13 depending on their performance. The first three tasks done were visual searches before the memory load was introduced after which the high and low load memory tasks were alternated with short breaks in between the trials.

After data analysis, the accurately yielded results on the memory task were generally high. The low-memory condition however, had a relatively higher result though. The results led to the conclusion that the cognitive tasks required depended on the memory task applied. In the visual search task, a high reward distractor resulted in an omission compared to when a low reward distractor appeared. The memory load also had an impact where participants were most likely to fixate on the distractors regardless of high or low value in the presence of a high memory load compared to when on a low memory load task. The interaction between the dense memory load and a lower memory load was significant though, with the VMAC effect being greater in the top load compared to the less load memory task. The influence of the reward on the attention capture was thus demonstrated to the enhanced when one had a high memory load. The visual search omission trials also revealed the effect of physical saliency on the proportion of omissions, with the exclusions being more likely in the presence of a coloured singleton than in its absence. Generally, under a high memory load, the effect of the physical saliency on attention capture was enhanced.

The second experiment aimed at assessing the replicability of the results obtained in the first experiment. The test administered was considered a to have a higher sensitivity value for the analysis of the effects of the load on the oculomotor capture stimulated by the presence of a reward. The trial involved the use of an occasional distractor containing both high value destructor and low value destructor, together in the view display so that they would both compete for attention from the participant. The researchers expected more eye movements to distractors when the participants were under top memory load as per the conclusion from the first experiment. Since both distractors had the same physical prominence, a dense memory load was anticipated to cause a distractor related eye movement towards the high reward distractor if at all memory load increased distraction by reward.

The study involved forty-three participants, each being paid a credit course of 30 Australian dollars and a bonus based on their performance. One participant was, however, excluded for having excess time outs. Orange was the top reward distractor colour for one part of the study population and blue, for the other. The study was formulated similarly to that in the first experiment one where the individuals first did three search tasks alone before memory load tasks were added in the following ten blocks. The trial featured ‘44 trials were done. 15 of them had only top reward destructors, 15 of low, 6 of no destructor identity and 8 with both.’ (Watson, Pearson, Chow et al, 2019). For the single distractors, the distractor was placed on one or two random locations away from the diamond, similar to the non-salient grey circle chosen to act as the exclusion causing stimulus. In the trial with both top and low reward distractors, each distractor placed as described above with the other set the same way but on the opposite side of the target. The reward points for both distractor trials was 500 points which were subject to omission when the gaze was detected on either of the distractors. The memory load was increased to six instead of five digits as in the first experiment. The ration of exclusion trial was still the dependent variable. The result of the memory task revealed that accuracy was significantly greater in the lesser-memory condition compared to the dense one. The VMAC effect was demonstrated by having more reward omissions oh high single distractor type than low single distractor type. As in the first experiment, the VMAC effect was heightened under a dense memory load. By analyzing the results from both high and low distractor type study, the above observation was supported. Competition between high and low-value distractors also yielded the under a high memory load, the distractor related eye movement was directed to the high reward distractor.

As this study tried to investigate if cognitive control can be used in preventing oculomotor capture by salient distractors, it was noted that sometimes the attention could be captured by physically salient distractors and this capture was also influenced by the presence of a reward. Supporting previous work done by Watson, Pearson, Chow et al. (2016), the results of this experiment showed that one’s attention was likely to be interfered with in the presence of a top reward destructor even when it meant a higher loss of the rewards. Moreover, both physical saliency and reward effects on oculomotor capture are magnified in conditions of high memory load. The second experiment served to show the counterproductive effect of reward on VMAC in the presence of a high memory load where participants had a discriminating increase in oculomotor capture of high reward distractors under such condition. The findings of this research answered the questions raised before, that the prioritization of distractors that are reward-related is not purely an automatic process but that it can be controlled and be reduced in the presence of sufficient cognitive resources. These findings hence support the hypothesis that working memory can provide selective visual attention to allow for minimal interference by goal irrelevant distractors in a goal-oriented task.

In conclusion, the implications of this research are essential, especially in the clinical context where they can be used to explain the reward-related intentional biases leading to compulsive behaviours as observed in cases of addictions. While previous studies had shown that the relationship between bias and substance abuse can be controlled by executive control in different individuals, this study shows the exact role of executive power in the relationship.  The behaviour where individuals (addicts) trying to abstain from drug and substance abuse are more likely to engage in drug behaviour in the presence of visual drug stimuli is hence accounted for by the presence of low cognitive control.

References

Pearson, D., Osborn, R., Whitford, T. J., Failing, M., Theeuwes, J., & Le Pelley, M. E. (2016). Value-modulated oculomotor capture by task-irrelevant stimuli is a consequence of early competition on the saccade map. Attention, Perception, & Psychophysics, 78, 2226–2240. doi:10.3758/s13414-016- 1135-2

Watson P., Pearson D., Chow M., Theeuwes J., Wiers R. W, Most S.B., LePelley M. (2019). Capture and Control: Working Memory Modulates Attentional Capture by Reward-Related Stimuli.