Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of UK Essays.
Learning and attention are a result of past knowledge and the salience of stimuli. Through the phenomenon of value-modulated attentional capture, cues that acquire salience or high-reward value are known to involuntarily capture attention, despite their detrimental effects on performance. However, it is unknown whether a reduction of reward expectancies in visual search paradigms can extinguish this bottom-up attentional bias. The present experiment assessed the impact of reversing participants’ beliefs about colour-value contingencies by manipulating reward instructions. Our findings demonstrate that value-modulated attentional capture is highly reward-dependent and that the effect can be extinguished when individuals’ reward expectancies are made redundant. This finding presents clinical implications for the population with substance-abuse and gambling disorders as treatment that is focused on reversal of cue-reward predictions may lower rates of relapse.
Learning and attention have long been thought of as a two-stage process whereby the organism first attends to stimuli in their environment, and then makes a behavioural response determined by the observation (Mackintosh 1975; Le Pelley, Johnson & Mitchell 2013). However, it is debated in the literature whether learning and attention are a result of top-down (voluntary) or bottom-up (involuntary) mechanisms (Anderson, Laurent & Yantis 2011; Anderson & Yantis 2013). Theeuwes (2010) conducted a review of brain studies including Functional Magnetic Resonance Imaging, which showed that initial attention is bottom-up as it is determined by the salience of a cue in the environment, and that only later can top-down conscious control occur. Salience refers to the perceptual qualities of stimuli that captivate an individual’s attention (Field & Cox 2008; Theeuwes 2010). For example, habitual poker-machine players find the sight of monetary symbols on screen, such as a pot of gold, salient, and thus capture attention to provoke repeated use (Shao, Read, Behrens & Rogers 2013).
Unfortunately, attention to such stimuli can be detrimental to maintaining goals (e.g. abstinence from gambling) and the learnt value of stimuli can be difficult to reverse (Anderson, Laurent & Yantis 2011; Le Pelley, Mitchell, Beesley, George & Wills 2016). Involuntary attention to stimuli is a result of prior learning that certain environmental cues (Mackintosh 1975) result in either high-value or low-value reinforcement, and high-value cues capture attention more than an equal cue with low-value reward; this phenomenon is known as value-modulated attentional capture (VMAC) (Anderson et al., 2013; Le Pelley, Pearson, Griffiths & Beesley 2015). Anderson et al., (2011) employed a visual search paradigm during the Experiment 1 training phase, whereby participants learnt to associate various coloured target-circles with high-or low-value monetary reward through inter-trial feedback; but at testing trials, the target became to identify the orientation of a bar within a diamond. Despite colour becoming ‘task-irrelevant’ at the test phase, reaction times were longer in trials where high-value colour distractors were present, hindering performance and reducing monetary gains. Therefore, the VMAC effect occurred automatically (Anderson et al., 2011). This effect persists long-term, up to seven to nine months after initial training (Anderson & Yantis 2013).
During Experiment 2 of Anderson et al., (2011), inter-trial reward feedback was withheld at training with new participants and the VMAC effect extinguished. This demonstrated that awareness of reward-value was necessary in VMAC contexts to manipulate the association of colour with value. In contrast, Le Pelley et al., (2015) conducted similar experiments to Anderson et al., (2011) but instead, training trials had the same instruction as test trials (attending to the line within a diamond) and colour-distractors merely signalled the reward that may be received on that trial. Therefore, the VMAC effect occurs even when participants are not consciously aware of the colour-reward association (Le Pelley et al., 2015). Importantly, rewards do not have to be monetary, as seen in Le Pelley et al. (2016)’s experiments where subjects performed for hypothetical alien currency; subjects learnt “Blargs” were worth more than “Gremps” and the VMAC effect occurred.
Field & Cox (2008) reviewed that those with substance-abuse disorders are more sensitive to attentional biases than those without, as a result drug-related stimuli involuntarily capture the attention of substance-abusers; this has devastating results as attentional capture by drug-related cues in the environment is predictive of relapse in substance-abusers (Anderson & Yantis 2013; Le Pelley et al., 2015). The literature has demonstrated that VMAC is not dependent upon participants awareness of colour-value contingencies, despite the results of Experiment 2 in Anderson et al., (2011). However, it is unknown whether an individual must have explicit expectations of receiving high and low-value rewards in order for a VMAC effect to occur. What if participants were instructed that colour-value contingencies became identical, such that they indicated the same level of reward? This presents a gap in the literature as VMAC may be dependent on reward-value expectancies.
Therefore, the current study hypothesised that the VMAC effect will occur for aware and unaware participants at training; high-value distractors will increase reaction times (RT) in both groups. It is also hypothesised that the VMAC effect will be extinguished when subjects are informed that trials are worth the same reward; mean RT will decrease in group 5.5 (all trials worth 5.5 points) but increase in group 10v1 (trials worth 10 points or 1 point).
The experimental design drew from Le Pelley et al. (2015)’s Experiments 1 and 2, however participants would attain imaginary points rather than money. Our study was mixed, utilising both a within-subjects and between-subjects design. Subjects completed the same practice and training phases, as well as an awareness test, equating to a within-subjects design. However, in the extinction phase participants were also randomly allocated to group “10v1” or group “5.5,” characteristic of a between-subjects design. Through this, the value of distractors could be manipulated such that each group in the extinction phase had different reward expectancies.
Practice phase. Participants were instructed to report the orientation of a line, vertical (press C) or horizontal (press M), inside a target diamond. The target was present among five circles, whereby there was one yellow distractor. Participants had a 2-second response deadline and trials continued until they reached 75% accuracy across 28 consecutive trials. No feedback was provided in-between trials regarding performance and no reward was presented.
Training phase. Similarly, to the practice phase, participants reported the lines orientation in the diamond on 300 trials. The response deadline was calculated for each individual as the 21st longest RT out of the last 28 trials in their practice block. However, 50% of trials featured one red circle distractor and 50% one green circle distractor. Colours were counterbalanced as signalling either high-value reward (10 points) or low-value reward (1 point). Feedback was provided in-between trials with the word “CORRECT” or “INCORRECT” displayed, and the corresponding reward for that trial shown as “10 points” or “1 point.”
Extinction phase. Prior to commencing the extinction phase, participants were randomly allocated to group 5.5 or group 10v1 and groups received separate instructions; either “all correct trials now worth 5.5 points” or “continue to receive 10 points or 1 point for correct trials” respectively. However, unlike the training phase, there were 120 trials and feedback was withheld from participants.
Awareness test. Upon completion of the extinction phase, participants were presented with 30 trials identical to the training phase, but they were also asked to report whether the trial would result in a 10 point or 1-point reward. The cut-off to be categorised as aware of the colour-value contingency was to score a minimum of 75% accuracy (23/30 trials) in reporting the expected reward. All other subjects were categorised into the unaware group as it is predicted they were responding at chance; they were unaware of the colour-value contingency.
Utilising the training phase data (see Figure 1), a 2 X 2 analysis of variance (ANOVA) revealed a significant main effect of distractor-value, F(1, 123) = 13.24, p < .001. However, the main effect of awareness, and the interaction effect between distractor-value and awareness, were non-significant, p > .05.
Figure 1. Mean reaction time (RT) across 300 practice phase trials for trials with high-value distractors and low-value distractors; participants either aware or unaware of colour-value contingency. RT is significantly slower for trials with high-value distractors present than low-value distractors. Error bars reflect within-subjects standard error of mean RT.
Utilising the extinction phase data for group 5.5 (see Figure 2), a 2 X 2 ANOVA yielded that the main effect of awareness and main effect of distractor-value were non-significant, p > .05. The interaction effect between awareness and distractor-value was also non-significant, p > .05.
Figure 2. Mean RT in group 5.5. across 120 extinction phase trials with high-value and low-value distractors; participants either aware or unaware of colour-value contingency. No significant differences in RT between aware or unaware participants when reward expectancy is 5.5 points per trial. Error bars reflect within-subjects standard error of mean RT.
In contrast, with group 10v1 in the extinction phase (see Figure 3), a 2 X 2 ANOVA yielded a significant interaction effect between distractor-value (high or low) and awareness (aware or unaware), F(1,60) = 4.89, p = .031. However, the main effect of awareness and the main effect of distractor-value were non-significant, p > .05.
Using a paired-samples t-test, the mean difference in RT between high-value and low-value distractors in the aware participants of group 10v1 was significant, t(1, 37) = 1.48, p = .027. However, the mean difference in RT between high-value and low-value distractors in the unaware participants of group 10v1 was non-significant, p > .05.
Figure 3. Mean RT in group 10v1 across 120 extinction phase trials with high-value and low-value distractors; participants either aware or unaware of colour-value contingency. Significant difference in mean RT between aware and unaware participants when reward expectancy is 10 or 1 points per trial. Error bars reflect within-subjects standard error of mean RT.
As hypothesised, the VMAC effect occurred in training regardless of participants’ awareness of the colour-value contingency; when a high-value distractor was present, participants took longer to respond to the target, indicating attentional-capture. Therefore, our experiment replicated the findings of Le Pelley et al., (2015) that VMAC did not depend on awareness of a colour-value contingency (Le Pelley et al., 2016).
The VMAC effect disappeared after extinction trials when participants’ reward expectancies changed; they were instructed that all correct trials resulted in the same reward (group 5.5) and participants were faster to respond to the target when all trials were made equal in the reward-sphere. In contrast, the VMAC effect was not extinguished when participants continued to expect high-reward and low-reward trials (group 10v1). Critically, this only occurred when participants were also aware of the colour-value contingency; unaware participants in this group extinguished the VMAC effect. Thus, our hypothesis that VMAC could be extinguished when reward expectancies changed was confirmed.
Attentional-capture did not occur for participants that were unaware of the association between distractor-colour and their signalling of high versus low-value reward. This strongly indicates that VMAC may be dependent on an individuals’ expectancy of receiving high versus low-value rewards, as well as awareness that specific cues in the environment signal reward. Our experiment presents significant findings to the literature on VMAC as attentional capture was driven by a conscious belief that the probability of receiving high-value rewards is more likely in one context than another.
Don & Livesey (2015) theorise attentional-capture as a product of Learnt Predictiveness, a phenomenon related to VMAC that organisms automatically refer to past knowledge to direct attention and guide behaviour. When an individual acquires knowledge that all cues are predictive of equal reinforcement, then the salience of such cues decreases (Mackintosh 1975) and top-down cognitive control can be exerted voluntarily to prevent distraction (Anderson et al., 2011). However, when it is acquired that certain stimuli are predictive of reinforcement, this knowledge is difficult to extinguish (Don & Livesey 2015); cues that were once predictive of high-value rewards are involuntarily attended to (Theeuwes 2010). Therefore, it is evident that VMAC is a result of bottom-up mechanisms.
Unfortunately, our experiment has a significant methodological flaw relating to the awareness test, which may negate our findings. An arbitrary cut-off of 75% accuracy was selected for inferring participants’ awareness of the colour-value contingency and it was assumed that participants who scored less than 75% were guessing rewards at chance. However, this is not a reliable measure of inferring awareness as participants were only exposed to 300 trials in the training phase in order to learn a colour-value contingency. It is not accurate to infer that the unaware participants were guessing rewards when they may have been subconsciously aware of the reward manipulation. Bourgeois, Neveu & Vuilleumier (2016) found that participants subconsciously attended to distractor-cues despite never reporting observing them; they were unaware that they were exposed to a coin for 16msec per trial. Therefore, it is invalid to assume participants were fully unaware of the colour-value contingency in our experiment.
Ultimately, our experimental finding that VMAC is dependent on reward-expectancy has clinical implications as substance-abusers are strongly influenced by stimuli in the environment that are predictive of reward (Field & Cox 2008). As stated previously, attentional-capture is maintained long-term even without exposure to reward stimuli (Anderson & Yantis 2013); therefore, repeated exposure to drug-related stimuli that does not result in rewarding drug-taking behaviour may be important and necessary to reverse reward-expectancies. Future research should endeavour to refine the experiment conducted in this study by assessing participant awareness of colour-value contingencies more reliably.
- Anderson, B.A., Laurent, P.A., & Yantis, S. (2011). Value-driven attentional capture. PNAS: Proceedings of the National Academy of Sciences of the United States of America, 108, 10367-10371.
- Anderson, B.A. & Yantis, S. (2013). Persistence of value-driven attentional capture. Journal of Experimental Psychology: Human Perception and Performance, 39, 6-9.
- Bourgeois, A., Neveu, R. & Vuilleumier, P. (2016). How does awareness modulate goal-directed and stimulus-driven shifts of attention triggered by value learning? PLOS One, 11, 1-13.
- Don, H.J. & Livesey, E.J. (2015). Resistance to instructed reversal of the learned Predictiveness effect. The Quarterly Journal of Experimental Psychology, 68, 1327-1347.
- Field, M. & Cox, W.M. (2008). Attentional bias in addictive behaviours: A review of its development, causes, and consequences. Drug and Alcohol Dependence, 97, 1-20.
- Le Pelley, M.E., Johnson, A.M. & Mitchell, C.J. (2013). Outcome value influences attentional biases in human associative learning: dissociable effects of training and instruction. Journal of Experimental Psychology: Animal Behaviour Processes, 39, 39-55.
- Le Pelley, M.E., Pearson, D., Griffiths, O., & Beesley, T. (2015). When goals conflict with values: counterproductive attentional and oculomotor capture by reward-related stimuli. Journal of Experimental Psychology, 144, 158-171.
- Le Pelley, M.E., Mitchell, C.J., Beesley, T., George, D.N., & Wills, A.J. (2016). Attention and associative learning in humans: An integrative review. Psychologica1 Bulletin, 142, 1111-1140.
- Mackintosh, N.J. (1975). A Theory of Attention: variations in the associability of stimuli with reinforcement. Psychological Review, 82, 276-298.
- Shao, R., Read, J., Behrens, T. & Rogers, R.D. (2013). Shifts in reinforcement signalling while playing slot-machines as a function of prior experience and impulsivity. Translational Psychiatry, 3, 1-9.
- Theeuwes, J. (2010). Top-down and bottom-up control of visual selection. Acta Psychologica, 135, 77-99.
If you need assistance with writing your essay, our professional essay writing service is here to help!Find out more
Cite This Work
To export a reference to this article please select a referencing style below:
Related ServicesView all
DMCA / Removal Request
If you are the original writer of this essay and no longer wish to have the essay published on the UK Essays website then please: