Exploring the Role of Effort and Impulsivity in Reversal Learning

Exploring the Role of Effort and Impulsivity in Reversal Learning

Exploring the Role of Effort and Impulsivity in Reversal Learning


The research hypothesized that decision-making abilities and dopamine (dopamine) are influential in reverse learning in people with varying levels of impulsivity. Preliminary research demonstrated that decision-making was influenced by a cost-benefit analysis when a positive or negative outcome was visibly apparent. In addition, dopamine and cortical brain regions were intricately connected in influencing reversal learning in Parkinson’s disease patients. The study involved 171 participants subjected to test using dynamometers to assess their probabilistic reverse learning and the UPPS-P instrument to measure their impulsivity. Inferential and descriptive statistics were used in data analysis. The study found that there was little difference between HiForce and LoForce values for the study sample. However, planning was positively associated with positive and negative urgency, showing that reward and decision-making were intricately connected with probabilistic reverse learning and impulsivity.   

Exploring the Role of Effort and Impulsivity in Reversal Learning

Impulsivity is a major factor in determining an individual’s personality. In addition, it is a major factor in many psychiatric disorders. It could predict an individual’s behavior, including binge shopping, susceptibility to addiction, and serial dating. In addition, rapid response impulsivity could be associated with human psychopathologies such as violent behavior. Most importantly, it could be related to biological factors such as dopamine, which is a neurotransmitter associated with the reward pathways and learning functions of the brain. Probabilistic reversal-learning paradigm has been essential in understanding the link between dopamine, personality, and effortful actions in learning. According to Clark, Cool, and Robbins (2004), “reversal learning involves the adaptation of behavior according to changes in stimulus-reward contingencies” (p. 45). Understanding how probabilistic reverse learning occurs in response to the environment can help deal with personality disorders associated with addictive tendencies and aggressive behaviors. The research hypothesizes that decision-making abilities and dopamine (reward) is influential in reverse learning in people with differing levels of impulsivity.

Research on probabilistic reversal learning in individuals with Parkinson’s disease could provide a better understanding of dopamine’s role in probabilistic reversal learning. Peterson et al. (2009) conducted research on how depletion of dopamine in patients with Parkinson’s affects their ability to adapt to reversed reward contingencies using probabilistic reward learning. The study consisted of 17 Parkinson’s disease patients who were off their dopamine medication and 15 participants of the same age (control group) to compare the effect of dopamine’s influence on probabilistic reward learning (Peterson et al., 2009). Peterson et al. (2009) found that the study sample had deficiencies in the post and pre-reversal tests on the ability to adapt to the reversal, which was dependent on the difficulty of the choices presented by the researchers. Furthermore, the researchers found that the patient’s perceived benefits based on pre-reversal feedback influenced the patients’ adaptability, thereby showing a clear link between reward (dopamine) and learning.

Further research on Parkinson’s disease shows that there could be pervasive motivational deficits when making incentivized decisions or those that involve a reward. Trevor et al. (2015) conducted novel research in which the study participants were asked about their willingness to squeeze a dynamometer. The researchers quantified their motivation by varying different magnitudes of reward with varying levels of squeezing. The researchers included an effort indifference point, which is the study participant’s reward acceptance probability of 50 percent. Similar to Peterson et al.’s (2009) study, the Trevor et al. (2015) study included control and study groups. The researchers tested the study group while they were on and off dopaminergic drugs to check whether dopamine influenced their response to rewards. The researchers found that Parkinson’s disease patients displayed lower efforts compared to controls regardless of the rewards. However, dopamine promoted effort allocation in patients with Parkinson’s disease, thereby reducing their decision-making deficits. Chong et al. (2015) have made similar findings regarding the positive influence of dopamine on Parkinson’s patients’ willingness to engage in a particular task. Neuroscience provides additional evidence for the link between decision-making, effort, and rewards in reverse learning.  

From a Neuroscience perspective, particular regions of the brain are responsible for reverse learning. Clark, Cool, and Robbins (2004) conducted a synthesis research on the neuropsychology of reversal learning and decision-making. The researchers found that dopamine neurotransmitter and ascending 5-HT systems had a modulator role in both processes, which validates the empirical research conducted on the link between reversal learning, decision-making, and effort in Parkinson’s disease patients. Consequently, Clark, Cool, and Robbins (2004) concluded that reversal learning and decision-making could be useful markers for studying the functional integrity of the ventral prefrontal cortex in neurological and psychiatric disorders. Additional research on the brain shows that the cortical regions are crucial in reverse learning and decision-making.

Cortical regions are associated with abstract thinking but their role in probabilistic reversal learning could be a substantial one. Cools et al. (2002) validated Clark, Cool, and Robbins’ (2004) linkage of decision-making and reverse learning using event-related functional magnetic resonance imaging. The researchers studied probabilistic reverse learning on 13 study participants. The probabilistic reverse learning task had stimulus-reward associations during the first performance of the task. The researchers introduced negative feedback to evaluate the regions of the brain that would be active during the probabilistic reversal-learning task. Similar to Clark, Cool, and Robbins (2004), the researchers found a significant change in the activity of the ventrolateral prefrontal cortex as the participants engaged in the probabilistic reverse learning task. The researchers found no impact on the signals in the aforementioned brain regions, including ventral stratum, when they introduced negative feedback (Cools et al., 2002). Other researchers have also found positive evidence to show that various cortical regions such as the medial prefrontal cortex and the orbitofrontal cortex are involved in reverse learning processes. In addition, they have found that lesions or damage to these regions leads to reversal learning deficits (Izquierdo et al., 2017; Clark, Cool, & Robbins, 2004). Cools et al.’s (2002) research suggest that negative feedback from the external environment might not be involved in the reverse learning process.



            The study participants were 171 college students aged between 22 and 25 years old. They were healthy adults who had no reported instances of mental health problems in the past since such disorders would lead to neurological disturbances such as chemical imbalances. Since the research demonstrated dopamine’s influence in probabilistic reverse learning processes, the sample had to have a relatively stable mental state to take part in the research.


            Two force dynamometers were used to measure the pressure with which the participants squeezed when presented with a stimulus. Two abstract shapes with varying levels of value were used to incentivize the study participants in performing the task. A modified version of the UPPS Impulsive Behavior Scale was used to measure learning and impulsivity relationships.


            The experiment was conducted in two stages. In the initial phase, the participants were required to squeeze the force dynamometer as hard as they could. Subsequently, they engaged in the learning task, in which the researcher provided them with two abstract shapes with a different value. The more and the less abstract shapes were rewarded 70 percent and 30 percent of the time. To ensure that the participants understood the more valuable stimuli, the researcher rewarded the more valuable stimuli with one point while the less valuable object received no point. To gauge the participants’ probabilistic reversal learning, the researcher periodically reversed the relative value of the study participants. The researcher asked the participants to re-evaluate their preferences when the reversal occurred, which requires the activation of reverse learning processes. The participants squeezed the left or right dynamometers when the researcher presented the stimuli to the right or left of their fixation. The results were collected in two blocks, one with a 5 percent maximum voluntary contraction (MVC) and one with 30 percent MVC. The researcher also manipulated the results by asking the study participants to squeeze on the dynamometer with a high or low force, after which the researcher simultaneously provided them with feedback.


            The research involved a descriptive and inferential approach to data analysis to answer the research questions. The independent variables were the personality traits measured using the UPPS-P scale while the dependent variables were HiForce and LoForce scores.


            The data was arranged in an excel database and imported into SPSS software. The descriptive and inferential analyses were conducted. Thereafter, the researcher paired the points scored with those on measures of impulsivity on the UPPS-P test to understand whether there were monotonic relationships between the points scored and impulsivity. The output was produced as PDF documents.

The descriptive statistics show that the means for the LoForce and HiForce were comparable with means of 98.86 and 97.07. The 5 percent trimmed mean, which results when the upper and lower 5 were deleted, and was very similar to the means of the HiForce (5 percent trimmed mean = 97.20 percent) and LoForce (5 percent trimmed mean = 98.78 percent), showing that there were few outliers in the values collected. Fewer outliers mean that the values collected in the research were consistent or had very few errors. The median values for the HiForce and LoForce were 97 and 99 respectively, showing little variation between the two variables. There was more variation in the maximum and minimum values than in the other descriptive statistics (Appendix I and II). The maximum values for the HiForce and LoForce were 114 and 118 respectively while the minimum values were 75 and 82 respectively. The HiForce had negative kurtosis and skewness values of -0.260 and -0.113, showing that the distribution of the data was lighter towards the tail end or left (Figure 2). On the other hand, LoForce had positive skewness values of 0.083, suggesting a relatively symmetrical distribution of values. The LoForce kurtosis value was -0.148, which shows that the values were lighter towards the tail end or left (Figure 1).

Figure 1: LoForce Histogram Showing Distribution of Values

Figure 2: HiForce Histogram Showing Distribution of Values

            The descriptive statistics for the accuracy of the HiForce and LoForce shows that there was an insignificant difference between the two variables. Figure 3 shows that the mean for the LoForce was 0.63, which was the same for the mean of the HiForce as well. There was a slight difference in the standard deviation, which was 0.064 for the LoForce and 0.068 for the HiForce, indicating that the two values had relatively consistent values. Additional descriptive statistics are provided in Appendix III and IV.

Figure 3: LoForce Histogram Showing Accuracy of Values

Figure 4: HiForce Histogram Showing Accuracy of Values

            The results from the correlations of scores recorded and impulsivity revealed that there was a moderate to strong correlation between planning and negative urgency (0.63). There was a moderate negative correlation between planning and positive urgency (-0.52). There was a weak to very weak correlation for negative and positive urgency for the perseverance measure of impulsivity and sensation seeking (Figure 5). The scale used to measure the weakness of the correlation coefficient in the values reported in Figure 5 is provided in Figure 6.

Figure 5: Correlations of Scores on the Test with Select Measures of Impulsivity

Figure 6: Scale for Determining the Strength of Pearson’s Correlation Coefficients


The results have demonstrated that there was little difference between the HiForce and LoForce values for the participants. The results show that the values were skewed towards the left of the histograms included in the results, showing that the study participants had low motivation to squeeze in both cases. In addition, the maximum and minimum values for the LoForce were relatively higher compared to those of HiForce, which shows that the study participants had a lower motivation to squeeze for the HiForce compared to LoForce because it takes less effort. Therefore, the reward is an essential component of a participant’s willingness to squeeze using the HiForce. Consequently, the HiForce had negative skewness and kurtosis values compared to LoForce, which only had a negative kurtosis value. The research conducted on Parkinson’s patients has demonstrated that reward is essential in ensuring that participants make the effort to engage in probabilistic reversal learning. The current research validates this contention and demonstrates that a cost-benefit analysis is an essential component of effort in probabilistic reversal learning. The research found little significant difference between the accuracy of the high force and low force blocks as indicated by the descriptive mean and standard deviation statistics.

The correlation between the questionnaire measure of impulsivity and the total number of points demonstrated that decision-making is crucial in engaging probabilistic reversal learning in the study participants. The only impulsivity measure that had a correlation with the scores was planning, which is related to decision-making. Higher scores were associated with a higher tendency towards planning. There was no relationship between perseverance and sensation seeking, and the scores in the study. This suggests that only particular measures of impulsivity are related to learning. The finding is consistent with the research on neuroscience, which has shown that brain regions associated with abstract thinking are associated with probabilistic reversal learning. The current research invalidated the finding in Cools et al.’s (2002) research that negative feedback did not influence probabilistic reversal learning. Consequently, the current research promotes the two-dimension theory of impulsivity as the framework that fits with the research findings.

A two-dimensional theory of impulsivity contends that impulsivity can be functional or dysfunctional based on how an individual processes information. In some contexts, it is more adaptive to act impulsively since this would be the most optimal pathway to adopt while in other contexts, responding impulsively could lead to negative outcomes. The two-dimensional theory supposes that forethought or the need to process information before making a decision has utility depending on contextual factors. Personality traits associated with functional impulsivity include being active, adventurous, and a high level of enthusiasm. On the other hand, impulsivity is associated with disorderliness, which is a tendency to make rash decisions by ignoring important facts during information processing (Whiteside & Lynam, 2001). The reason other measures of impulsivity did not return correlations with the scores could be that impulsivity is a concept that “subsumes several moderately related constructs that play different roles in accounting for risky behavior” (Cyders et al., 2015, p. 107). Future research should attempt to narrow down the definition and make it more precise to ensure consistency in the conceptual framework on impulsivity. Decision-making and reward should appear concurrently in future research since the personality and neuropsychological research suggests that the two are inextricably linked. The major limitation of the current research is that it does not include neurological data to understand the internal processes of the study participants as they engaged in the tasks. Future research should provide study participants with and without mental health issues with complex tasks to gauge differences in their decision-making processes. Insights developed from such research could help understand how rewards can be manipulated to influence an individual’s probabilistic reversal learning in clinical contexts to reduce impulsive behavior in people with mental health problems.


Chong, T. T. J., Bonnelle, V., Manohar, S., Veromann, K. R., Muhammed, K., Tofaris, G. K., …

& Husain, M. (2015). Dopamine enhances willingness to exert effort for reward in

Parkinson’s disease. cortex, 69, 40-46.

Clark, L., Cools, R., & Robbins, T. W. (2004). The neuropsychology of ventral prefrontal cortex:

decision-making and reversal learning. Brain and cognition, 55(1), 41-53.

Cools, R., Clark, L., Owen, A. M., & Robbins, T. W. (2002). Defining the neural mechanisms of

probabilistic reversal learning using event-related functional magnetic resonance imaging. Journal of Neuroscience, 22(11), 4563-4567.

Cyders, M. A., Smith, G. T., Spillane, N. S., Fischer, S., Annus, A. M., & Peterson, C. (2007).

Integration of impulsivity and positive mood to predict risky behavior: development and validation of a measure of positive urgency. Psychological assessment, 19(1), 107.

Izquierdo, A., Brigman, J. L., Radke, A. K., Rudebeck, P. H., & Holmes, A. (2017). The neural

basis of reversal learning: an updated perspective. Neuroscience, 345, 12-26.

Peterson, D. A., Elliott, C., Song, D. D., Makeig, S., Sejnowski, T. J., & Poizner, H. (2009).

Probabilistic reversal learning is impaired in Parkinson’s disease. Neuroscience, 163(4),


Whiteside, S. P., & Lynam, D. R. (2001). The five factor model and impulsivity: Using a

structural model of personality to understand impulsivity. Personality and individual differences, 30(4), 669-689.


  1. Descriptive Statistics: HiForce and LoForce
  1. Descriptive Statistics: HiForce and LoForce
  1. Descriptive Statistics: Accuracy of HiForce and LoForce Values
  1. Descriptive Statistics: Accuracy of HiForce and LoForce Values
Calculate your order
Pages (275 words)
Standard price: $0.00
Client Reviews
Our Guarantees
100% Confidentiality
Information about customers is confidential and never disclosed to third parties.
Original Writing
We complete all papers from scratch. You can get a plagiarism report.
Timely Delivery
No missed deadlines – 97% of assignments are completed in time.
Money Back
If you're confident that a writer didn't follow your order details, ask for a refund.

Calculate the price of your order

You will get a personal manager and a discount.
We'll send you the first draft for approval by at
Total price:
Power up Your Academic Success with the
Team of Professionals. We’ve Got Your Back.
Power up Your Study Success with Experts We’ve Got Your Back.
The Custom Writing