Boosting cognitive control: Does the timing of positive feedback make a difference? Elizabeth Uduwa-vidanalage
11392339
Internship Project Supervisor: Dr. Nils Jostmann
Abstract
Cognitive control pertains to our ability to regulate thoughts and actions based on goals or specific tasks. Further understanding the mechanisms that facilitate this process could help us enhance cognitive performance. Previous research has indicated that positive affect appears to be a key factor in the regulation of cognitive control but mixed findings have been accumulated. Whereas some studies have revealed a facilitative effect, others show a dampening influence. The purpose of the present study was to investigate whether the timing of positive feedback, a source of positive affect, plays a crucial role in its influence on cognitive control. Using a multi trial Stroop task, positive feedback was presented either before a trial (pre-goal) or immediately after a trial performance (post-goal). Based on previous research (Kazen & Kuhl, 1999), we predicted that there would be a significant interaction effect between feedback timing (pre-goal and post-goal) and valence (positive and neutral) favoring pre-goal positive feedback. The initial results revealed a non-significant interaction effect however further analyses demonstrated promising findings supporting pre-goal positive feedback (i.e. smaller Stroop interference) over post-goal positive feedback, pre-goal neutral feedback and post-goal neutral feedback.
Furthermore, given that people may differ in their sensitivity to rewards (i.e. how rewarding they may find the positive feedback), the influence of individual differences of BAS-reward
responsiveness on Stroop interference and positive feedback was also examined. The present findings bring new insight into the psychology of performance motivation by investigating positive affect and the conditions under which it is most beneficial.
Keywords: positive, feedback, affect, cognitive control, pre-goal
Boosting cognitive control: Does the timing of positive feedback make a difference? The human mind is capable of remarkable things, one of which is the ability to regulate thoughts and actions based on internal goals or specific tasks. This ability is commonly referred to as cognitive control. Investigating cognitive control provides us with insight on how to enhance our cognitive capacities and performance. To date, various studies have focused on the structure and actual influence of cognitive control (Banich, 2009; Botvinick, Braver, Barch, Carter, & Cohen, 2001; Miller, 2000), but little is known about its underlying mechanisms. Nevertheless, from the studies that have been conducted so far it has become relatively clear that related brain systems (i.e. prefrontal cortex) are involved both in cognitive control processing and emotion and motivation processing (Chiew & Braver, 2011; Ray & Zald, 2012). Thus, not surprisingly, recent studies have illustrated that emotions appear to modulate cognitive control processes in terms of its content as well as its strength. For example, positive emotions and moods have been linked with greater flexible processing such as an extended scope of attention or increased shifting between cognitive sets (Goschke & Bolte, 2014).
In line with these findings, positive affect, any short-lived or long-lasting experience of pleasure that could be either conscious or unconscious, appears to play a key factor in the regulation of cognitive control. Various studies have revealed that positive affect facilitates a wide-range of cognitive processes (Ashby, Isen & Turken, 1999; Isen, 2001 review).
Nevertheless, most of the studies have focused on the facilitation of positive affect on a variety of problem-solving and decision making tasks in general but not directly on its effect on cognitive control per se (Dreisbach & Goschke, 2004). The results of the few studies that do, reflect mixed findings. While there is evidence that suggests that positive mood hinders
performance on central executive control tasks (Oaksford, Morris, Grainger, Williams, & Mark 1996), other studies suggest that positive affect appears to facilitate cognitive control (Kazen & Kuhl, 2005; Kuhl & Kazen, 1999).
More specifically, Kuhl and Kazen (1999) explored the influence of phasically induced positive affect (i.e. positive primes) on the classic Stroop task. The Stroop task involves a difficult intention, the intention to respond to the color in which the color word is written in, while overcoming the stronger tendency to actually read the color word (i.e. if red is written in blue ink, the intention is to say blue but the competing tendency is to read red). Typically, participants take longer to respond to incongruent stimuli (i.e. red in blue ink) than to control
stimuli (XXXX’s in blue) and this is known as the Stroop interference effect. Based on the authors’ predictions, these intentions, or planned actions, are kept active in a higher order, symbolic form in intention memory, the system involved in the regulation between intentions and execution, when there is a series of actions or steps to take place. They are presumably kept there until the appropriate moment comes to respond; until then the respond is inhibited. Based on their model, positive affect may play a role in the release of this inhibition, signaling that the right moment has arrived. Their results revealed that indeed positive affect led to a form of volitional facilitation (decreased Stroop interference), but only when there were at least two sequential Stroop tasks per trial (series of steps). This finding appears rational given that if there was only one step, monitoring of sequencing performed through a central executive system (leading to the release of the inhibition) would not be required. Not surprisingly, the facilitative effect was not present for the second trial (possibly because there was only one step left.
Furthermore, in one of their follow up studies Kazen and Kuhl (2005 study 4) investigated whether there was an alternative method to activate intentions. In other words, they sought to load intention memory on a trial by trial basis by forming an “achievement-related context” in the start of each trial (p. 440). This was accomplished by providing participants with feedback related to their performance before each trial. Their results demonstrated that in fact
achievement-related positive primes stimulated volitional facilitation even when there was only one Stroop task per trial, thus indicating that the achievement-related positive primes were sufficient to load intention memory.
However, the question still remains as to why positive affect (i.e. positive feedback) does not always have beneficial effects (Oaksford et al., 1996). Perhaps, the timing of positive affect plays an essential role in whether it is constructive or not. More specifically, various theories suggest that positive affect may have a different effect on cognitive processes when it is
experienced in anticipation of performance as opposed to in response to performance (Gable & Harmon-Jones, 2011). By default, feedback is usually provided after performance (post-goal) however there is evidence to suggest that post-goal feedback sometimes appears to have a dampening effect on performance or else referred to as “coasting” (Carver, 2003; Gable & Harmon-Jones, 2011). More specifically in this case, positive affect appears to demonstrate that resources can be allocated to a different task or goal because satisfactory performance has been attained.
On the other, Custers and Aarts (2010) argued that when positive affect (i.e. positive word) is experimentally associated to a mental goal it facilitates subsequent performance by allocating effort towards the completion of the goal. This argument comes in line with Kuhl and Kazen’s (1999) finding that positive primes presented before performance led to increased cognitive control. Furthermore, Jostmann (2016; unpublished) found that positive feedback, a source of positive affect, framed as pre-goal feedback facilitated subsequent performance in a number search task in comparison to post-goal feedback. In other words, feedback presented immediately after the participant’s performance (post-goal) was not as beneficial as feedback provided after the announcement of the subsequent task (pre-goal). Hence, whereas pre-goal positive affect seems to assist succeeding performance, post-goal positive affect may jeopardize it.
The purpose of the present study was to shed light into the optimal timing of feedback by examining whether positive affect framed as pre-goal feedback would lead to volitional
facilitation in comparison to post-goal feedback in the classic Stroop task. Prior to the task(s), an announcement (i.e. “1 task” or “1+2 tasks”) appeared informing participants what would follow. Post-goal feedback was presented immediately after performance of the first trial whereas pre-goal feedback was presented after the announcement of the subsequent trial. The announcement was an explicit way to elicit intention memory by informing the participants of the number of steps or actions that would follow and to segregate post-goal feedback (before announcement) from pre-goal feedback (after announcement). It was predicted that positive feedback, controlled for neutral feedback, provided immediately after the announcement (pre-goal) would lead to greater volitional facilitation (i.e. removal/decreased Stroop interference effect) in comparison to feedback provided before the announcement (post-goal).
Additionally, another factor that was investigated because of its potential role in the effect of positive affect is the individual difference in reward responsiveness. According to Gray there are two major systems regulating behavioral activity, the behavioral activation system (BAS) and the behavioral inhibition system (BIS; Gray 1971, 1982, as cited in Carver & White, 1994). The former is thought to be sensitive to signals of reward whereas the latter is perceived to be sensitive to signals of punishments. Hence, individuals with higher BAS levels would be expected to be behaviorally more sensitive to reward signals (i.e. proneness to engage towards goal-directed efforts) and to experience more positive affect, than people with lower levels of BAS as a result (Krupic & Corr, 2014). Hence it was expected that volitional facilitation could
be stronger for individuals with higher BAS; specifically, BAS-reward responsiveness was investigated as a moderator of the expected valence x timing interaction.
Method Participants
Participants were recruited from an online research portal of the University of Amsterdam and were primarily students from the University of Amsterdam or Hogeschool of Amsterdam. Overall, a sample of 90 participants started and completed the experiment and received either 1 research credit or 10 euros for their participation. From this sample, 12 participants were excluded from the analyses for the following reasons: one because of color blindness, one because of possible intoxication, eight because they failed to understand the nature of feedback (i.e. misunderstood pre-goal feedback), and two because their error rate in the Stroop task was more than 2.5 SDs above the group mean (M = 5.02%; SD = 12.50). As a result, 78 participants remained (M = 24.60; range from 18 to 60), of which 59 were female. All instructions and materials (including an informed consent) were presented to participants in Dutch and the study was approved by the ethics committee of the University of Amsterdam’s psychology department (project code 2017-SP-7681).
Materials
BAS reward-responsiveness. The BIS/BAS scale is comprised of a 24-item self-report questionnaire that is used to assess personality differences defined by variation in BAS and BIS sensitivities (Carver & White, 1994; see Appendix 1). Given that participants were provided with positive feedback (a form of reward) a prediction was made only for one subscale of BAS, the reward responsiveness scale. The subscales fun-seeking and drive do not tackle reward
sensitivity per se and thus were not included in the confirmatory analysis; also since no explicit negative feedback was presented no predictions were made for the BIS scale either. As for the reward-responsiveness scale it determines the tendency to feel positive affect in the presence of reward cues and it consists of 5 items (i.e. “When I’m doing well at something, I love to keep at it). Participants had to respond on a 4-point Likert scale ranging from 1 (very true to me) to 4 (very false for me). After reverse scoring the items higher scores indicated higher levels of BAS-reward responsiveness. Moreover, the BIS/BAS is a well-known scale and its psychometric properties have been widely accepted portraying a high internal reliability (Franken, Muris, &
Rassin, 2005; Johnson, Turner, & Iwata, 2003; Poythress et al., 2008). For the present study, Cronbach’s alpha for the BAS reward-responsiveness subscale was 0.70.
Procedure
Participants initially provided some demographic information and completed the
BIS/BAS questionnaire before starting the Stroop task. Prior to the initiation of the Stroop task, participants were provided with the following information. They learned that some trials
consisted of a single task and some of two consecutive tasks and that before the trial they would be notified if they had to do only one task ("Task 1") or two tasks ("Task 1 + Task 2"). They also learned that they would receive feedback after a trial regarding their performance but that they would not always receive feedback. They would get positive feedback (smiley face) if they performed well on the previous trial (they responded faster in the previous trial compared to the average of the last 10 trials). If they did not receive feedback, they would only see a grey circle. The grey circle could mean two things: (a) they have responded faster to the previous trial than the average of the last 10 trials but the computer has determined that they will not see any feedback or (b) they have responded to the previous trial exactly as well or slower than the average of the last 10 trials. It was stressed to the participants that if they did not get feedback (grey circle), it may still be that they responded faster but if they received positive feedback (smiley face) that they definitely reacted faster (than the average of last 10 trials). Finally, they were informed that feedback would be presented to them sometimes directly after their
performance and sometimes after the announcement of the next trial because in real life feedback is usually provided when people are already involved in something else. This was emphasized to ensure that participants always understood that the feedback was regarding their performance on the previous trial.
They began with 8 warm-up trials to familiarize themselves with the buttons, during which they received accuracy feedback (“Correct” or “Incorrect”). Then they completed 16 practical trials without accuracy feedback before they started the main trials. The main trials were divided into two blocks. In the first block, apart from the 128 main trials there were 16 additional trials in the beginning that were not analyzed (to ensure that participants had understood the experiment’s flow and pace). After a small break, participants completed the second block which consisted of another 128 trials. Overall, they were presented with a total of 256 main trials. Upon completion of the Stroop task they had to fill a manipulation check
questionnaire (measuring their attention, motivation, and comprehension of instructions; see Appendix C). All questionnaires were created and presented using Qualtrics. The whole experiment lasted approximately 60 minutes (see Figure 1 for a visual overview of the experiment).
The Stroop task was designed using the Presentation program and all instructions were presented on screen (see Appendix 2). The participants’ task was to name the color of the presented stimulus (either a word or a row of XXXX) by pressing on the corresponding key as quickly and accurately as possible. The correct answer at all times was either blue, green, red, or yellow and there was a colored sticker on each of the keys (“A”, “S”, “K”, and “L”) to avoid confusion. The order of the key colors was counterbalanced.
Regarding the technical aspects of the Stroop task, there was a preprogrammed equal distribution of the number of tasks in each trial (1 or 2), pre-goal and post-goal feedback, and incongruent and control stimuli, with the order of presentation being random for each participant. Furthermore, feedback was performance contingent. That is, positive feedback was based on the individual reaction times of the last 10 accurate responses (for incongruent and control trials separately). When participants were faster than the 75th percentile (the mean of the 7th and 8th RT when the RTs were arranged in an ascending order), 66.6% of the time they received positive feedback (a smiley face). They received neutral feedback (grey circle) the remaining 33.3% of the time and when they were slower than the 75th percentile. As a result, neutral feedback was not necessarily negative but it was rather uninformative. They only received explicit negative feedback (“INCORRECT”) when they made an error. Overall, there was an approximately equal distribution of positive (42.88%) and neutral feedback (49.55%) and a small percentage of negative feedback (7.57%; see Table 1 for detailed frequency across conditions).
Moreover, the SOA (stimulus onset asynchrony) between feedback and the first Stroop task was held constant at 2250ms. In the post-goal feedback trials, feedback was presented for 750ms followed by the announcement (1500ms); while for the pre-goal feedback trials, the announcement was presented for 1500ms followed by feedback (750ms) and a blank screen (1250ms; see Figure 2). Previous research has indicated that positive primes presented for 750ms to 2250ms have the same effect on Stroop performance (Kazen & Kuhl, 2005).
Design
In summary, the following study followed a 2 (positive vs. neutral) × 2 (pre-goal vs. post-goal) × 2 (single task vs. double tasks) × 2 (control vs. incongruent) within-subjects design with cognitive control (i.e. Stroop interference in response times) as the dependent variable.
Results Data preparation
For each participant, RTs were removed if they were 2.5 standard deviations above or below their mean. The remaining RTs were converted to natural logarithms in order to ensure that the normality of data is met (typically done in RT experiments). Subsequently, mean RTs were calculated for each participant per condition. Finally, Stroop interference indices were obtained by subtracting RTs of control conditions from the corresponding incongruent conditions. The reported statistics were obtained using natural logarithms but for the ease of interpretation means and standard deviations were reported in milliseconds. Alpha was set at 0.05.
Main analyses
A 2 (feedback timing: pre-goal vs post-goal) x 2 (feedback valence: positive vs. neutral) x 2 (trial type: single task vs. double tasks) repeated measures ANOVA was conducted with Stroop interference (in response times) on the first Stroop task of each trial as the dependent variable. The results revealed a significant main effect of trial type on Stroop interference, F (1, 77) = 5.01, p = .028, η2 =.061. Specifically, Stroop interference was smaller in the first task of double task trials (M = 32, SE = 5) in comparison to single task trials (M = 45, SE = 6). Additionally, a marginally significant effect of timing was detected, F (1, 77) =3.62, p=.061, η2=.045. That is, Stroop interference was smaller when pre-goal feedback (M=34, SE=5) was provided in
comparison post-goal feedback (M=44, SE=6). Moreover, a significant interaction effect between valence and trial type was also revealed, F (1, 77) =5.78, p=.019, η2=.070 (see Figure 3).
Unraveling this interaction indicated that after receiving positive feedback, Stroop interference was smaller in the first task of double task trials (M=27, SE=7) than in single task trials (M=51, SE=8), p= .004. Conversely, after receiving neutral feedback there was no difference in Stroop interference between double task trials (M=39, SE=7) and single task trials M=37, SE=6), p= .945. Furthermore, there seemed to be no difference in Stroop interference after positive
However, there was a marginally significant difference in Stroop interference after receiving positive feedback (M= 27, SE= 7) compared to after receiving neutral feedback (M= 39, SE= 7) in double task trials, p= .076. Regarding the main effect of valence, p=.818, and the interaction effects between timing and trial type, p=.494, and timing, valence, and trial type, p=.964, they all yielded nonsignificant results.
Main a-priori hypothesis. The main a priori prediction was that we would find a significant interaction effect between valence and timing. Contrary to expectations, the 2 (feedback timing: pre-goal vs post-goal) x 2 (feedback valence: positive vs. neutral) x 2 (trial type: single task vs. double tasks) repeated measures ANOVA which was conducted with Stroop interference (in response times) on the first Stroop task of each trial as the dependent variable revealed a non-significant interaction effect between valence and timing, p=.258.
Exploratory analyses
Because the effect of positive affect on Stroop interference was only detected in double task trials, only data from the first task of double task trials (and not of single task trials) were used to conduct the following analysis. The 2 (feedback timing: pre-goal vs. post-goal) x 2 (feedback valence: positive vs. neutral) repeated measures ANOVA which was conducted with Stroop interference as the dependent variable revealed a marginally significant effect of timing, F(1,77)=3.45, p=.067, η2=.043. More specifically, Stroop interference was smaller after the administration of pre-goal feedback (M=25, SE=6) as opposed to post-goal feedback (M=40, SE=7). Similarly, a marginally significant main effect of valence was also detected F(1,77)=3.22, p=.076, η2=.040. Particularly, Stroop interference was smaller after receiving positive feedback (M=27, SE=7) compared to neutral feedback (M=37, SE=6). However, the interaction effect between timing and valence was not significant, p=.406.
Nevertheless, pairwise comparisons with contrasts revealed interesting results trending in the direction we expected (see Figure 4). That is, Stroop interference was marginally smaller after receiving pre-goal positive feedback (M=17, SE=8) in comparison to post-goal positive feedback (M=37, SE=10), p= .059 and pre-goal neutral feedback (M= 33, SD=8), p =.068. On the contrary, the difference between pre-goal neutral feedback (M=33, SE=8) and post-goal neutral feedback (M=42, SE=9) was insignificant similarly to the difference between post-goal neutral feedback (M=42, SE=9) and post-goal positive feedback (M=37, SE=10) which was also insignificant, p=.411.
BAS reward-responsiveness
A 2 (feedback timing: pre-goal vs. post-goal) x 2(feedback valence: positive vs. neutral) repeated measures ANCOVA was conducted with BAS-reward (centered) as the covariate and Stroop interference (first task in double task trials) as the dependent variable. The analysis revealed that BAS-Reward was unrelated to the dependent variable, F(1,78)=0.02, p=.895. Since a relationship between the covariate and the dependent variable is required for an ANCOVA to be sensible it would be meaningless to report and interpret further aspects of the analysis. Exploratory analyses of moderators
Various exploratory moderators were examined to investigate whether they affected the results. For example, differences in participants’ level of interest in the task, subjective difficulty of the task, motivation to participate, and attention were all variables that could affect the results hence they were examined as possible moderators. The 2 (feedback timing: pre-goal vs post-goal) x 2 (feedback valence: positive vs neutral) x 2 (trial type: single task vs double task) repeated measures ANCOVA with Stroop interference as the dependent variable and interest, difficulty, and motivation as centered covariates revealed no significant results, p >.05.
Additionally, we investigated whether there were significant differences between the first block and the second block. To explore this, a 2 (feedback timing: pre-goal vs post-goal) x 2 (feedback valence: positive vs neutral) x 2 (trial type: single task vs double task) x 2 (block: first vs second) repeated measures ANOVA was conducted with Stroop interference as the dependent variable. The results indicated that there was no significant difference between the two blocks and there was no interactive effect either, p > 0.05.
Discussion
The main objective of the present study was to investigate the effect of positive feedback, framed as a pre-goal source of positive affect, on cognitive control. More specifically, it was expected that pre-goal positive feedback, controlled for neutral feedback, would facilitate cognitive control in comparison to post-goal feedback. Pre-goal positive affect was
operationalized by providing feedback after the information of the subsequent trial (i.e. single or double task) was presented but before actual performance. Theoretically, the information about the trial (i.e. number of steps that they need to take) is thought to activate intention memory and positive affect in turn can facilitate the execution of the intention (Kuhl & Kazen, 1999). Hence, if positive feedback is provided directly after the activation of the intention (i.e. after
announcement) this should be especially beneficial for succeeding performance. However contrary to our predictions, we did not detect a statistically significant interaction effect between feedback valence and timing.
Nevertheless, further exploratory inspections revealed promising results in favor of pre-goal positive feedback. That is, pre-pre-goal positive feedback led to volitional facilitation (i.e. decreased Stroop interference) in comparison to post-goal positive feedback. It could be argued at this point that the timing per se and not the combination of pre-goal and positive (feedback) resulted in increased cognitive control. However, no difference was detected in Stroop
interference between pre-goal neutral feedback and post-goal neutral feedback. Also, when compared to pre-goal neutral feedback, pre-goal positive feedback led to decreased Stroop interference. These results are quite promising and may offer a possible explanation for the discrepancies typically found in research related to positive feedback (Kazen & Kuhl, 2005; Kuhl & Kazen, 1999; Oaksford et al., 1996). In other words, the timing of positive feedback may play a role in the nature of its effect (i.e. positive vs. negative influence). Notably, the effect of post-goal positive feedback did not differ from post-goal neutral feedback thus failing to
emphasize the detrimental coasting effect that has been suggested by some studies (Carver, 2003; Gable & Harmon-Jones, 2011); nevertheless, this still implies that post-goal feedback is no different than no feedback (i.e. neutral) as opposed to pre-goal positive feedback’s advantageous nature. Nonetheless, it is important to stress that these results are merely lower level descriptive information that depict encouraging findings but further research is required before we draw sturdy conclusions.
Moreover, another interesting finding that surfaced was that the aforementioned results were only detected in double task trials and not in single task trials. Although no explicit prediction was made for double vs single trials, based on previous research we anticipated that there would be no difference between types of trials (single vs double task trials). The rationale underlying this assumption was based on Kazen and Kuhl’s (2005; study 4) finding that an achievement related context (i.e. positive primes) formed in the beginning of each trial would be sufficient to activate intention memory. Nonetheless, our results do not appear to confirm this finding since positive feedback (specifically pre-goal) facilitated cognitive control only when there were two consecutive tasks, thus confirming Kuhl and Kazen’s (1999) initial finding that a series of (difficult) steps or actions are required in order to activate intention memory and in turn
for positive affect to help execute the intention into an action. In other words, the fact that the type of trial x valence interaction was significant could be a posteriori interpreted as a
(conceptual) replication of Kuhl and Kazen (1999).
Alternatively, a possible explanation as to why we found the (marginal) effect only in single task trials may be that we were unsuccessful in activating intention memory through the means of an achievement related context because of our neutral feedback’s description. To clarify, Kazen and Kuhl (2005; study 4) presented positive, neutral, and negative primes each one indicating specific information about their performance (performed better, performed the same, performed worse than a standard) but in the present experiment neutral feedback was uninformative. In other words, participants received neutral feedback either when they
performed the same or worse than their average but also 30% of time they performed better. So, participants knew that even if they presented well they might not receive positive feedback. This may have influenced the achievement related context we were trying to create thus resulting to different results than Kazen and Kuhl (study 4; 2005).
It could also be argued that the uninformative nature of feedback may be responsible for the weak findings (of comparisons) that were mentioned above. Specifically, although pre-goal positive feedback led to decreased Stroop interference in comparison to post-goal positive feedback and pre-goal neutral feedback, the statistical significance of these results was rather weak. Perhaps this was due to the interpretation each participant provided for the neutral feedback. For example, if participants viewed neutral feedback as positive (i.e. they performed well but the computer did not provide them with feedback) its effect would not be different from the effect of positive feedback. Put differently, some neutral feedback may have actually carried on the weight of positive feedback thus “polluting” the statistical differentiation between positive and neutral feedback.
Moreover, various exploratory analyses were conducted to further investigate the data. To illustrate, task difficulty, motivation, and interest were explored as moderators to ensure that differences in these variables did not significantly change the results. However, none of the variables appeared to play a prominent role in the findings. Additionally, it could be reasoned that differences in the effect of positive feedback could be detected between the first block and the second block. For example, participants may have found the Stroop task and the feedback exciting and challenging in the beginning but the repetitive nature of the task may have caused
them to lose interest in the feedback towards the second half of the Stroop task. To explore this possibility, we investigated whether there was a significant difference between the two blocks, however the results failed to detect such a difference thus suggesting that the effect of positive feedback was similar across blocks.
Finally, the only moderator where an a priori prediction was made was BAS-reward, where participants with higher levels of BAS-reward were expected to benefit the most from positive feedback because of their sensitivity to rewards. Contrary to expectations, BAS-reward did not have a significant relationship to Stroop interference and it did not moderate the expected volitional facilitation (i.e. decreased Stroop interference). Perhaps the BAS-reward subscale did not reflect the participants´ actual levels of reward responsiveness given that self-report
questionnaires are usually susceptible to social desirability. Alternatively, maybe participants underestimated or overestimated their responsiveness to reward, as can sometimes be the case with explicit measurements of personality (Asendorpf, Banse, & Mücke, 2002). Nonetheless, it would be interesting for future studies to investigate whether more implicit measurements of reward responsiveness could actually moderate the relationship between positive affect and cognitive control.
Overall, the present study revealed some interesting findings concerning positive affect and cognitive control. That is, positive affect experienced in anticipation of performance (pre-goal feedback) appeared to be more beneficial than positive affect experienced as a response of performance (post-goal). Interestingly, this beneficial effect of pre-goal feedback was only detected when there was a series of tasks to complete. In other words, a series of tasks was required for intention memory to load and for positive feedback to facilitate the pathway between the intention and its execution. These results come in line with previous findings examining the activation of intention memory and its enactment (Kuhl & Kazen, 1999), and may be a possible solution to the contradicting results regarding positive affect found in the literature. That is, the timing of feedback appears to play a role in the effectiveness of feedback. Nevertheless, it would be rather premature to draw robust conclusions given the strength of the present results. Future studies are encouraged to further explore these discoveries by comparing positive with negative feedback and by investigating positive feedback in comparison to different forms of neutral feedback (i.e. informative vs. uninformative). Feedback is an important aspect of our lives and it
is usually provided to help us improve performance, thus understanding the factors that could tamper with its advantageous character is vital.
References
Ashby, F. G., Isen, A. M, & Turken, U. (1999). A neuropsychological theory of positive affect and its influence on cognition. Psychological review, 106(3), 529-550.
Asendorpf, J. B., Banse, R., & Mücke, D. (2002). Double dissociation between implicit and explicit personality self-concept: The case of shy behavior. Journal of Personality and
Social Psychology, 83(2), 380-393.
Atkinson, J. W. (1957). Motivational determinants of risk-taking behavior. Psychological
Review, 64, 359-372.
Banich, M. T. (2009). Executive function: The search for an integrated account. Current
Directions in Psychological Science, 18(2), 89-94.
Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, C. S., & Cohen, J. D. (2001). Conflict monitoring and cognitive control. Psychological Review, 108(3), 624-652.
Carver, C. (2003). Pleasure as a sign you can attend to something else: Placing positive feelings within a general model of affect. Cognition & Emotion, 17(2), 241-261.
Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: The BIS/BAS Scales. Journal of
Personality and Social Psychology, 67(2), 319-33.
Custers, R., & Aarts, H. (2010). The unconscious will: How the pursuit of goals operates outside of conscious awareness. Science, 329(5987), 47-50.
Chiew, K. S., & Braver, T. S. (2011). Positive affect versus reward: emotional and motivational influences on cognitive control. Frontiers in Psychology, 2, 1-10.
Dreisbach, G., & Goschke, T. (2004). How positive affect modulates cognitive control: Reduced perseveration at the cost of increased distractibility. Journal of Experimental
Psychology-Learning Memory and Cognition, 30(2), 343-352.
Franken, I. H., Muris, P., & Rassin, E. (2005). Psychometric properties of the Dutch BIS/BAS scales. Journal of Psychopathology and Behavioral Assessment, 27(1), 25-30.
Gable, P. A., & Harmon-Jones, E. (2011). Attentional consequences of pregoal and postgoal positive affects. Emotion, 11(6), 1358-1367.
Goschke, T., & Bolte, A. (2014). Emotional modulation of control dilemmas: the role of positive affect, reward, and dopamine in cognitive stability and flexibility. Neuropsychologia, 62, 403-423.
Gray, J. A. (1972). The psychophysiological basis of introversion extraversion: A modification of Eysenck’s theory. In V. D. Nebylitsyn & J. A Gray (Eds.) The biological bases of individual behavior. San Diego: Academic Press. 182–205.
Gray, J. A. (1981). A critique of Eysenck’s theory of personality. In a model for personality.
Springer Berlin Heidelberg. 246-276
Isen, A. M. (2001). An influence of positive affect on decision making in complex situations: Theoretical issues with practical implications. Journal of Consumer Psychology, 11(2), 75-85.
Johnson, S. L., Turner, R. J., & Iwata, N. (2003). BIS/BAS levels and psychiatric disorder: An epidemiological study. Journal of Psychopathology and Behavioral Assessment, 25(1), 25-36.
Kazén, M., & Kuhl, J. (2005). Intention memory and achievement motivation: Volitional facilitation and inhibition as a function of affective contents of need-related stimuli. Journal of Personality and Social Psychology, 89(3), 426.
Krupić, D., & Corr, P. J. (2014). Individual differences in emotion elicitation in university examinations: A quasi-experimental study. Personality and Individual Differences, 71, 176-180.
Kuhl, J., & Kazén, M. (1999). Volitional facilitation of difficult intentions: joint activation of intention memory and positive affect removes stroop interference. Journal of
Experimental Psychology: General, 128(3), 382.
Miller, E. K. (2000). The prefontral cortex and cognitive control. Nature Reviews Neuroscience,
1(1), 59-65.
Oaksford, M., Morris, F., Grainger, B., & Williams, J, Mark. G. (1996). Mood, reasoning, and central executive processes. Journal of Experimental Psychology: Learning, Memory,
and Cognition, 22(2), 476.
Poythress, N. G., Skeem, J. L., Weir, J., Lilienfeld, S. O., Douglas, K. S., Edens, J. F., &
Kennealy, P. J. (2008). Psychometric properties of Carver and White’s (1994) BIS/BAS scales in a large sample of offenders. Personality and Individual Differences, 45(8), 732-737.
Ray, R. D., & Zald, D. H. (2012). Anatomical insights into the interaction of emotion and cognition in the prefrontal cortex. Neuroscience & Biobehavioral Reviews, 36(1), 479-501.
Table 1
Frequency of each Condition
Condition 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Percentage
(%)
5.6 5.3 5.4 5.3 5.3 5.4 5.6 5.4 6.3 6.3 6.0 6.2 6.3 6.0 6.0 6.4 7.2
Note. 1= Control, One task, Positive Pre-Goal Feedback; 2= Control, One task, Positive Post-Goal Feedback; 3= Control, Two tasks, Positive Pre-Post-Goal Feedback; 4= Control, Two tasks, Positive Post-Goal Feedback; 5= Incongruent, One task, Positive Pre-Goal
Feedback; 6= Incongruent, One task, Positive Post-Goal Feedback; 7= Incongruent, Two tasks, Positive Pre-Goal Feedback; 8= Incongruent, Two tasks, Positive Post-Goal Feedback; 9= Control, One task, Neutral Pre-Goal Feedback; 10= Control, One task, Neutral Post-Goal Feedback; 11= Control, Two tasks, Neutral Pre-Goal Feedback; 12= Control, Two tasks, Neutral Post-Goal Feedback; 13= Incongruent, One task, Neutral Pre-Goal Feedback; 14= Incongruent, One task, Neutral Post-Goal Feedback; 15= Incongruent, Two tasks, Neutral Pre-Goal Feedback; 16= Incongruent, Two tasks, Neutral Post-Goal Feedback; 17=Trials with Negative Feedback
Figure 1. An overview of the experiment’s procedure.
Figure 2. Example of trials. On the left we see a sample trial with 1 task and pre-goal positive feedback. Whereas on the right we see a sample trial with 2 tasks and post-goal neutral feedback.
Figure 3. The interaction effect between valence and type of trial on Stroop interference in response times (i.e. incongruent response times – control response times).
Figure 4. Pairwise comparisons indicating the differences in Stroop interference in response times between feedback valence and timing in double task trials.
Appendix A
BIS/BAS Questionnaire and Scoring
Each item of this questionnaire is a statement that a person may either agree with or disagree with. For each item, indicate how much you agree or disagree with what the item says. Please respond to all the items; do not leave any blank. Choose only one response to each statement. Please be as accurate and honest as you can be. Respond to each item as if it were the only item. That is, don't worry about being "consistent" in your responses. Choose from the following four response options:
1 = very true for me 2 = somewhat true for me 3 = somewhat false for me 4 = very false for me
1. A person's family is the most important thing in life.
2. Even if something bad is about to happen to me, I rarely experience fear or nervousness. 3. I go out of my way to get things I want.
4. When I'm doing well at something I love to keep at it.
5. I'm always willing to try something new if I think it will be fun. 6. How I dress is important to me.
7. When I get something I want, I feel excited and energized. 8. Criticism or scolding hurts me quite a bit.
9. When I want something I usually go all-out to get it.
10. I will often do things for no other reason than that they might be fun. 11. It’s hard for me to find the time to do things such as get a haircut. 12. If I see a chance to get something I want I move on it right away.
13. I feel pretty worried or upset when I think or know somebody is angry at me. 14. When I see an opportunity for something I like I get excited right away. 15. I often act on the spur of the moment.
16. If I think something unpleasant is going to happen I usually get pretty "worked up." 17. I often wonder why people act the way they do.
18. When good things happen to me, it affects me strongly.
19. I feel worried when I think I have done poorly at something important. 20. I crave excitement and new sensations.
21. When I go after something I use a "no holds barred" approach. 22. I have very few fears compared to my friends.
23. It would excite me to win a contest.
24. I worry about making mistakes. --- Items other than 2 and 22 are reverse-scored.
BAS Drive: 3, 9, 12, 21
BAS Fun Seeking: 5, 10, 15, 20
BAS Reward Responsiveness: 4, 7, 14, 18, 23 BIS: 2, 8, 13, 16, 19, 22, 24
Appendix B
Instructions for Stroop task
Uitleg 1
Het doel van dit onderzoek is om te weten te komen hoe mensen op feedback reageren tijdens een prestatietaak.
We leggen je nu uit hoe dit onderzoek is opgebouwd. Lees alle instructies zorgvuldig door. Je gaat nu een test doen. De test duurt in totaal ongeveer 30 minuten. Op de helft van de test zal er een korte pauze zijn.
Door op SPACE te drukken kan je telkens verder gaan naar het volgende scherm. [page break]
Uitleg 2
Elke trial bestaat uit de zogenaamde Stroop taak die er als volgt uitziet:
Er verschijnt een kleurwoord (bijvoorbeeld GROEN of ROOD) of een reeks van XXXX op het scherm.
De kleurwoorden of de reeks van XXXX worden in een bepaalde kleur weergegeven, bijvoorbeeld ROOD geschreven in groene kleur.
Je moet per druk op een knop aangeven in welke kleur het woord of de reeks van XXXX is weergegeven. Je mag niet reageren op de betekenis van het kleurwoord. Die moet je negeren. Wanneer je bijvoorbeeld GROEN geschreven in blauw ziet, druk je zo snel mogelijk op de toets op het toetsenbord met de blauwe sticker erop.
Wanneer je bijvoorbeeld XXXX geschreven in rood ziet, druk je zo snel mogelijk op de toets op het toetsenbord met de rode sticker erop.
Probeer zo snel en zo accuraat mogelijk te werken! [page break]
Uitleg 3
Sommige trials bestaan uit een enkele Stroop taak.
Bij andere trials moet je twee Stroop taken achter elkaar doen.
Vóór de trial krijg je een melding of je slechts één taak moet doen ("taak 1") of twee taken ("taak 1 + taak 2").
[page break] Uitleg 4
Je krijgt soms ná een trial feedback over hoe je hebt gepresteerd. Maar je krijgt niet altijd feedback te zien.
Je krijgt positieve feedback als je op de voorafgaande trial goed gepresteerd hebt. Dat betekent dat je sneller hebt gereageerd op de voorafgaande trial dan gemiddeld op de 10 trials daarvóór. Positieve feedback wordt weergegeven door een smiley.
betekenen:
1) misschien heb je sneller gereageerd op de voorafgaande trial dan gemiddeld op de 10 trials daarvóór. Je hebt dus goed gepresteerd maar de computer heeft bepaald dat je toch geen feedback te zien krijgt.
2) misschien heb je even snel of zelf langzamer gereageerd op de voorafgaande trial dan gemiddeld op de 10 trials daarvóór. In dat geval krijg je nooit feedback te zien.
Dus als je geen feedback krijgt (en dus de witte cirkel verschijnt), kan het nog steeds zo zijn dat je sneller gereageerd hebt. Als je wel feedback krijgt (en de smiley ziet) dan weet je zeker dat je sneller gereageerd hebt.
[page break] Uitleg 5
Als je feedback krijgt, dan kan het zijn dat je die direct na afloop van een trial te zien krijgt. Soms krijg je de feedback echter pas later te zien nadat je hebt gelezen of de volgende trial uit een Stroop taak (“taak 1”) of uit twee Stroop taken (“taak 1 + taak 2”) bestaat. De reden hiervoor is dat je in het echte leven soms pas feedback krijgt als je alweer bezig bent met iets anders.
[page break]
We beginnen met enkele oefentrials om je aan de juiste toetsen te laten wennen. Daarna komen er enkele “warming up” trials zodat je een gevoel voor de Stroop taak ontwikkelt. Daarna beginnen de echte trials.
Leg je wijsvingers en je middelvingers op de toetsen met een gekleurde stikker. Zorg ervoor dat je handgewrichten comfortabel op het tafelblad liggen.
Als je met de duim op SPACE drukt beginnen de oefentrials. Tijdens de oefentrials krijg je te zien of je wel of niet de juiste toets hebt gedrukt.
Je krijgt echter nog geen feedback over je snelheid. Succes!
[page break]
[after the first trials]
Dit waren de oefentrials. We gaan nu verder met de warming up trials.
TER HERINNERING: sommige trials bestaan uit slechts uit één Stroop taak ("taak 1"), en andere trials bestaan uit twee Stroop taken achter elkaar ("taak 1 + taak 2").
Je krijgt vóór de trial een melding uit hoeveel taken de trial bestaat. Tijdens de warming up trials krijg je nog geen feedback over je snelheid. Druk op SPACE om met de warming up trials te beginnen.
[page break]
Dit waren de warming up trials. Als je op SPACE drukt ga je verder met de echte trials. TER HERINNERING:
- als je een verkeerde knop hebt gedrukt krijg je “INCORRECT” te zien. Als je de juiste knop hebt gedrukt, krijg je soms wel en soms geen feedback.
- Als je WEL feedback krijgt (een smiley) betekent het dat sneller bent dan je eigen gemiddelde snelheid op de 10 trials daarvoor.
Dus je kunt alleen feedback krijgen als je goed hebt gepresteerd.
- Als je GEEN feedback krijgt (een witte cirkel), kan het betekenen dat je goed óf dat je slecht gepresteerd hebt.
[page break] PAUZE!
Je hebt nu 1 minuut pauze om je even te strekken. Druk daarna op SPACE om verder te gaan. [page break]
Dit was de laatste trial. Je bent klaar met de Stroop taak. Ga nu naar de proefleider. Hij of zij zal je nog enkele vragen voorleggen over hoe je deze test hebt ervaren.
Appendix C
Manipulation Check Questionnaire
