Dissociable after-effects of prosocial acts: Effort is costly for others but valued for self
Curation statements for this article:-
Curated by eLife
eLife Assessment
This paper provides solid electrophysiological evidence that an individual's effort expenditure increases the subjective value of a subsequent reward when the beneficiary is the individual themselves, but decreases the subjective value of a reward when the beneficiary is someone else. These findings have valuable implications for our understanding of how effort investment shapes reward evaluation during prosocial behavior.
This article has been Reviewed by the following groups
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
- Evaluated articles (eLife)
Abstract
Engaging in prosocial behavior requires effort, yet people are often averse to exerting effort for others’ benefit. However, it remains unclear how effort exertion affects subsequent reward evaluation during prosocial acts. Here, we combined high-temporal-resolution electroencephalography with a paradigm that independently manipulated effort and reward for self and others to elucidate the neural mechanisms underlying the reward after-effect of prosocial effort expenditure. We found dissociable reward after-effects for self-benefiting and other-benefiting effort. For self-benefiting rewards, the reward positivity (RewP) increased with effort demand, suggesting an effort-enhancement effect. In contrast, for other-benefiting rewards, the RewP decreased as effort increased, demonstrating an effort-discounting effect. Critically, this dissociation was contingent upon high reward magnitude and modulated by individual differences in effort discounting, yet remained distinct from performance evaluation. Our findings reveal distinct neural computations for self- and other-benefiting efforts, offering new insights into how prior effort expenditure shapes reward evaluation during prosocial behavior.
Article activity feed
-
-
-
eLife Assessment
This paper provides solid electrophysiological evidence that an individual's effort expenditure increases the subjective value of a subsequent reward when the beneficiary is the individual themselves, but decreases the subjective value of a reward when the beneficiary is someone else. These findings have valuable implications for our understanding of how effort investment shapes reward evaluation during prosocial behavior.
-
Reviewer #1 (Public review):
Summary:
The authors test the hypotheses, using an effort-exertion and an effort-based decision-making task, while recording brain dynamics with EEG, that the brain processes reward outcomes for effort differentially when they earned for themselves versus others.
Strengths:
The strengths of this experiment include what appears to be a novel finding of opposite signed effects of effort on the processing of reward outcomes when the recipient is self versus others. Also, the experiment is well-designed, the study seems sufficiently powered, and the data and code are publicly available.
Weaknesses:
There is some concern about the fact that participants report feeling less subjective effort, but also more disliking of tasks when they were earning rewards for others versus self. The concern is that participants …
Reviewer #1 (Public review):
Summary:
The authors test the hypotheses, using an effort-exertion and an effort-based decision-making task, while recording brain dynamics with EEG, that the brain processes reward outcomes for effort differentially when they earned for themselves versus others.
Strengths:
The strengths of this experiment include what appears to be a novel finding of opposite signed effects of effort on the processing of reward outcomes when the recipient is self versus others. Also, the experiment is well-designed, the study seems sufficiently powered, and the data and code are publicly available.
Weaknesses:
There is some concern about the fact that participants report feeling less subjective effort, but also more disliking of tasks when they were earning rewards for others versus self. The concern is that participants worked with less vigor during self-versus-others trials and this may partly account for a key two-way Recipient x Effort interaction on the size of the Reward Positivity EEG component. Of note, participants took longer to complete tasks when working for others. While it is true that, in all cases, participants met the requisite task demands (they pressed the required number of buttons) they did so more sluggishly when earning rewards for others. The Authors argue that this reflects less motivation when working for others, which is a plausible explanation. The Authors also try to rule out this diminished vigor as a confounding explanation by showing that the two way interaction remains even when including reaction times (and also self-reported task liking) as a covariate. Nevertheless, it is possible that covariates do not fully account for the effects of differential motivation levels which would otherwise explain the two-way interaction. As such, I think a caveat is warranted regarding this particular result.
-
Reviewer #2 (Public review):
Summary:
Measurements of the reward positivity, an electrophysiological component elicited during reward evaluation, have previously been used to understand how self-benefitting effort expenditure influences processing of rewards. The present study is the first to complement those measurements with electrophysiological reward after-effects of effort expenditure during prosocial acts. The results provide solid evidence that effort adds reward value when the recipient of the reward is the self but discounts reward value when the beneficiary is another individual.
Strengths:
An important strength of the study is that amount of effort, the prospective reward, the recipient of the reward, and whether the reward was actually gained or not were parametrically and orthogonally varied. In addition, the researchers …
Reviewer #2 (Public review):
Summary:
Measurements of the reward positivity, an electrophysiological component elicited during reward evaluation, have previously been used to understand how self-benefitting effort expenditure influences processing of rewards. The present study is the first to complement those measurements with electrophysiological reward after-effects of effort expenditure during prosocial acts. The results provide solid evidence that effort adds reward value when the recipient of the reward is the self but discounts reward value when the beneficiary is another individual.
Strengths:
An important strength of the study is that amount of effort, the prospective reward, the recipient of the reward, and whether the reward was actually gained or not were parametrically and orthogonally varied. In addition, the researchers examined whether the pattern of results generalized to decisions about future efforts. The sample size (N=40) and mixed-effects regression models are also appropriate for addressing the key research questions. Those conclusions are plausible and adequately supported by statistical analyses.
-
Author response:
The following is the authors’ response to the original reviews.
Public Reviews:
Reviewer #1 (Public review):
The authors test the hypotheses, using an effort-exertion and an effort-based decision-making task, while recording brain dynamics with EEG, that the brain processes reward outcomes for effort differentially when they earned for themselves versus others.
The strengths of this experiment include what appears to be a novel finding of opposite signed effects of effort on the processing of reward outcomes when the recipient is self versus others. Also, the experiment is well-designed, the study seems sufficiently powered, and the data and code are publicly available.
We thank Reviewer #1 for the affirmative appraisal of our manuscript as well as the thoughtful and insightful comments, which have enabled us to …
Author response:
The following is the authors’ response to the original reviews.
Public Reviews:
Reviewer #1 (Public review):
The authors test the hypotheses, using an effort-exertion and an effort-based decision-making task, while recording brain dynamics with EEG, that the brain processes reward outcomes for effort differentially when they earned for themselves versus others.
The strengths of this experiment include what appears to be a novel finding of opposite signed effects of effort on the processing of reward outcomes when the recipient is self versus others. Also, the experiment is well-designed, the study seems sufficiently powered, and the data and code are publicly available.
We thank Reviewer #1 for the affirmative appraisal of our manuscript as well as the thoughtful and insightful comments, which have enabled us to significantly improve the manuscript.
(1) Inferences rely heavily on the results of mixed effects models which may or may not be properly specified and are not supported by complementary analyses.
We thank Reviewer #1 for raising this critical issue of model specification. We have re-fitted our mixed-effects models and performed complementary analyses to validate the robustness of our findings. Specifically, we adopted the maximal converging random-effects structure (including random slopes for Recipient, Effort, and Magnitude where feasible) while ensuring model stability (see Responses to Reviewer #1’s Recommendations point 2). Crucially, our primary findings, including the Recipient × Effort and Recipient × Effort × Magnitude interactions, remained robust. Furthermore, additional analyses confirmed that these results were not confounded by factors such as response speed and subjective effort rating (see Responses to Reviewer #1’s Recommendations point 5).
(2) Also, not all results hang together in a sensible way. For example, participants report feeling less subjective effort, but also more disliking of tasks when they were earning rewards for others versus self. Given that participants took longer to complete tasks when earning effort for others, it is conceivable that participants might have been working less hard for others versus themselves, and this may complicate the interpretation of results.
We thank Reviewer #1 for this insightful point (which also relates to Reviewer #3’s point 5). In our study, participants were asked to rate three specific dimensions: Effort (“How much effort did you exert to complete each effort condition when earning rewards for yourself [or the other person]?”), Difficulty (“How much difficulty did you perceive in each effort condition when earning rewards for yourself [or the other person]?”), and liking (“How much did you like each effort condition when earning rewards for yourself [or the other person]?”).
We acknowledge the Reviewer #1’s concern that the lower subjective effort ratings for others seems contradictory to the higher disliking and longer completion times. We propose that in this paradigm, subjective effort ratings are susceptible to demand characteristics and likely captured motivational engagement (e.g., “how hard I tried” or “how willing I was”) rather than perceived task demands. To disentangle these factors, we included a measure of perceived task difficulty, which is anchored in task properties and is less prone to social desirability biases (Harmon-Jones et al., 2020; Wright et al., 1990). We found no differences in perceived difficulty between self- and other-benefiting trials (Figure 2D), suggesting that the task demands were perceived as equivalent across conditions. To examine this interpretation more directly, we analyzed correlations among participants’ ratings of difficulty, effort, and liking. As illustrated in Figure S1, we found no correlation between difficulty and effort ratings. Crucially, liking ratings were negatively correlated with difficulty ratings.
More importantly, our performance data contradict the interpretation that participants “worked less hard” for others in terms of task completion. While participants took longer to complete tasks for others, they maintained comparable, near-ceiling success rates for self (97%) and other (96%) recipients (b = -0.46, p = 0.632; Supplementary Table S1). This dissociation suggests that although participants were less motivated (e.g., lower subjective ratings, longer completion times, and greater disliking) to work for others, they ultimately exerted the necessary physical effort to achieve successful outcomes. Thus, the results consistently point to a decrease in prosocial motivation (consistent with prosocial apathy) rather than a failure of effort exertion.
Wright, R. A., Shaw, L. L., & Jones, C. R. (1990). Task demand and cardiovascular response magnitude: Further evidence of the mediating role of success importance. Journal of Personality and Social Psychology, 59(6), 1250-1260. https://doi.org/10.1037/0022-3514.59.6.1250
Harmon-Jones, E., Willoughby, C., Paul, K., & Harmon-Jones, C. (2020). The effect of perceived effort and perceived control on reward valuation: Using the reward positivity to test a dissonance theory prediction. Biological Psychology, 107910. https://doi.org/10.1016/j.biopsycho.2020.107910
Reviewer #2 (Public review):
Measurements of the reward positivity, an electrophysiological component elicited during reward evaluation, have previously been used to understand how self-benefitting effort expenditure influences the processing of rewards. The present study is the first to complement those measurements with electrophysiological reward after-effects of effort expenditure during prosocial acts. The results provide solid evidence that effort adds reward value when the recipient of the reward is the self but discounts reward value when the beneficiary is another individual.
An important strength of the study is that the amount of effort, the prospective reward, the recipient of the reward, and whether the reward was actually gained or not were parametrically and orthogonally varied. In addition, the researchers examined whether the pattern of results generalized to decisions about future efforts. The sample size (N=40) and mixed-effects regression models are also appropriate for addressing the key research questions. Those conclusions are plausible and adequately supported by statistical analyses.
We appreciate Reviewer #2’s positive appraisal of our manuscript. We are fortunate to receive your thoughtful and insightful suggestions and have revised the manuscript accordingly.
(1) Although the obtained results are highly plausible, I am concerned whether the reward positivity (RewP) and P3 were adequately measured. The RewP and P3 were defined as the average voltage values in the time intervals 300-400 ms and 300-440 ms after feedback onset, respectively. So they largely overlapped in time. Although the RewP measure was based on frontocentral electrodes (FC3, FCz, and FC4) and the P3 on posterior electrodes (P3, Pz, and P4), the scalp topographies in Figure 3 show that the RewP effects were larger at the posterior electrodes used for the P3 than at frontocentral electrodes. So there is a concern that the RewP and P3 were not independently measured. This type of problem can often be resolved using a spatiotemporal principal component analysis. My faith in the conclusions drawn would be further strengthened if the researchers extracted separate principal components for the RewP and P3 and performed their statistical analyses on the corresponding factor scores.
We thank Reviewer #2 for raising this issue. We would like to clarify that these two components were time-locked to different types of feedback and therefore reflect neural responses to distinct stages of the prosocial effort task. Specifically, the P3 was time-locked to performance feedback (the effort-completion cue; e.g., the tick shown in Figure 1B), whereas the RewP was time-locked to reward feedback (e.g., the display of “+0.6”). Thus, despite the numerical similarity in the post-stimulus windows, the components capture neural activity evoked by independent events separated in time, corresponding to the performance monitoring versus reward evaluation stages of the task. To avoid misunderstanding, we have made this distinction more explicit in the revised manuscript, which now reads, “Single-trial RewP amplitude was measured as mean voltage from 300 to 400 ms relative to reward feedback onset (i.e., reward delivery) over frontocentral channels (FC3, FCz, FC4). We also measured the parietal P3 (300–440 ms; averaged across P3, Pz, and P4) in response to performance feedback (i.e., effort completion), given its relationship with motivational salience (Bowyer et al., 2021; Ma et al., 2014)” (page 27, para. 1, lines 2–6).
Reviewer #3 (Public review):
This study investigates how effort influences reward evaluation during prosocial behaviour using EEG and experimental tasks manipulating effort and rewards for self and others. Results reveal a dissociable effect: for self-benefitting effort, rewards are evaluated more positively as effort increases, while for other-benefitting effort, rewards are evaluated less positively with higher effort. This dissociation, driven by reward system activation and independent of performance, provides new insights into the neural mechanisms of effort and reward in prosocial contexts.
This work makes a valuable contribution to the prosocial behaviour literature by addressing areas that previous research has largely overlooked. It highlights the paradoxical effect of effort on reward evaluation and opens new avenues for investigating the mechanisms underlying this phenomenon. The study employs well-established tasks with robust replication in the literature and innovatively incorporates ERPs to examine effort-based prosocial decision-making - an area insufficiently explored in prior work. Moreover, the analyses are rigorous and grounded in established methodologies, further enhancing the study's credibility. These elements collectively underscore the study's significance in advancing our understanding of effort-based decision-making.
We thank Reviewer #3 for the positive assessment. We are particularly encouraged by the reviewer’s recognition of our novel integration of ERPs to uncover the distinct effects of effort on reward evaluation for self versus others. We have carefully addressed the specific recommendations raised in the subsequent comments to further strengthen the rigor and clarity of the manuscript.
(1) Incomplete EEG Reporting: The methods indicate that EEG activity was recorded for both tasks; however, the manuscript reports EEG results only for the first task, omitting the decision-making task. If the authors claim a paradoxical effect of effort on self versus other rewards, as revealed by the RewP component, this should also be confirmed with results from the decision-making task. Omitting these findings weakens the overall argument.
We thank Reviewer #3 for giving us the opportunity to verify the specific roles of our two tasks. The primary aim of our study is to elucidate the neural after-effects of effort exertion on subsequent reward evaluation during prosocial acts. The prosocial effort task was specifically designed for this purpose, as it involves actual effort expenditure followed by reward outcomes. Furthermore, this task uses preset effort-reward combinations, ensuring balanced trial counts and adequate signal-to-noise ratios across conditions, a critical requirement for robust ERP analysis. In contrast, the prosocial decision-making task was included specifically to quantify behavioral preference (i.e., prosocial effort discounting) rather than neural reward processing. Specifically, this task involves choices without immediate effort execution and reward feedback, making it impossible to examine the neural after-effects of effort exertion. However, the decision-making task remains indispensable for our study structure: it provides an independent behavioral phenomenon of prosocial apathy, which allowed us to link individual differences in behavioral motivation to the neural dissociations observed in the prosocial effort tasks (as detailed in our Responses to Reviewer #3’s 2). Thus, the two tasks provide complementary, rather than redundant, insights into the behavioral and neural mechanism of prosocial effort.
(2) Neural and Behavioural Integration: The neural results should be contrasted with behavioural data both within and between tasks. Specifically, the manuscript could examine whether neural responses predict performance within each task and whether neural and behavioural signals correlate across tasks. This integration would provide a more comprehensive understanding of the mechanisms at play.
We thank Reviewer #3 for this insightful and helpful suggestion. We agree that linking neural signatures with behavioral patterns is crucial for establishing the functional significance for our ERP findings. Regarding within-task association, it is important to note that the prosocial effort task was designed to require participants to exert fixed, preset levels of physical effort to earn uncertain rewards. This experimental control was necessary to standardize effort exertion across self-benefiting and other benefiting trials, thereby minimizing confounds such as differences in physical or perceived effort prior to the feedback phase. Indeed, the neural after-effects remained after controlling for these behavioral measures (i.e., response speed and self-reported effort; as detailed in responses to Reviewer #1’Recommendations point 5). Furthermore, unlike the prosocial effort task, the decision-making task inherently precludes the examination of the neural after-effects of effort; therefore, within-task association in this task was not possible.
Given these considerations, we focused on the cross-task association. We examined whether the neural after-effects of effort (indexed by the RewP) in the prosocial effort task were modulated by individual differences in effort discounting. We used the K value estimated from the prosocial decision-making task as the index of effort discounting. We entered the K value (log-transformed and z-scored) as a continuous predictor into the mixed-effects models of RewP amplitudes. The full regression estimates for the model are presented in Table S1 (left).
We observed a significant four-way interaction among recipient, effort, magnitude, and K value (b = 0.58, p = 0.013). To decompose this complex interaction, we performed simple slopes analyses separately for self- and other-benefiting trials at high and low levels of reward magnitude and discounting rate (±1 SD). As shown in Figure S2, for self-benefiting trials, the effort-enhancement effect on the RewP was significant only for participants with high discounting rates at low reward magnitude (b = 1.02, 95% CI = [0.22, 1.82], p = 0.012). In contrast, participants with low discounting rates exhibited no significant effort effect (b = -0.37, 95% CI = [-0.89, 0.15], p = 0.159). At high reward magnitude, simple slopes analyses detected no significant effort effects for either high (b = 0.35, 95% CI = [-0.44, 1.14], p = 0.383) or low (b = 0.45, 95% CI = [-0.07, 0.97], p = 0.093) discounting individuals. These findings strongly support the cognitive dissonance account (Aronson & Mills, 1959): those who find effort most aversive are most compelled to inflate the value of small rewards to justify their exertion. For these individuals, the completion of a costly action for a small reward may trigger a stronger internal justification effect, resulting in an amplified neural reward response.
For other-benefiting trials, participants with low discounting rates exhibited a significant effort-discounting effect at high reward magnitude (b = -0.97, 95% CI = [-1.74, -0.20], p = 0.014). In contrast, no significant effort effects were observed for participants with high discounting rates at either high (b = -0.45, 95% CI = [-0.97, 0.08], p = 0.098) or low (b = -0.16, 95% CI = [-0.69, 0.38], p = 0.564) reward magnitudes, nor for participants with low discounting rates at low reward magnitude (b = 0.14, 95% CI = [-0.64, 0.92], p = 0.729). These results suggest that the justification mechanism observed for self-benefiting effort appears absent for other-benefiting effort. Instead, we observed a persistent effort discounting before, during, and after effort expenditure, which was most pronounced in individuals with low effort sensitivity (low K) when reward magnitude was high. This seemingly paradoxical pattern might be interpreted through the lens of disadvantageous inequity aversion (Fehr & Schmidt, 1999). Specifically, the combination of high personal effort and high monetary reward for another person creates a salient disparity between the participant’s incurred cost and the recipient’s gain. Although low-K individuals are behaviorally willing to tolerate this cost, their neural valuation system may nonetheless track the “unfairness” of this asymmetry, thereby attenuating the neural reward signal (Tricomi et al., 2010). These insights suggest that facilitating prosocial behavior may require not just lowering costs, but potentially framing outcomes to trigger the effort justification mechanisms that drive the effort paradox observed in self-benefiting acts (Inzlicht & Campbell, 2022).
To confirm this four-way interaction, we also replaced the high-effort choice proportions in the decision-making task and observed a similar four-way interaction among recipient, effort, magnitude, and high-effort choice proportions (b = -0.58, p = 0.014; see Table S1 for detailed regression estimates). Together, this cross-task analysis not only provides a more comprehensive understanding of the mechanisms at play but also justifies the inclusion of the prosocial decision-making task. We sincerely thank Reviewer #3’ for this valuable suggestion, which has significantly strengthened our manuscript. We have included this analysis (page 16, para. 2; page 17, paras. 1–2) and discussed the results (page 20, para. 2, lines 10–15; page 20, para. 3; page 21, para. 1, lines 1–8) in the revised manuscript.
Aronson, E., & Mills, J. (1959). The effect of severity of initiation on liking for a group. The Journal of Abnormal and Social Psychology, 59(2), 177-181. https://doi.org/10.1037/h0047195
Fehr, E., & Schmidt, K. M. (1999). A theory of fairness, competition, and cooperation. The Quarterly Journal of Economics, 114(3), 817-868. http://www.jstor.org/stable/2586885
Tricomi, E., Rangel, A., Camerer, C. F., & O'Doherty, J. P. (2010). Neural evidence for inequality-averse social preferences. Nature, 463(7284), 1089-1091. https://doi.org/10.1038/nature08785
(3) Success Rate and Model Structure: The manuscript does not clearly report the success rate in the prosocial effort task. If success rates are low, risk aversion could confound the results. Additionally, it is unclear whether the models accounted for successful versus unsuccessful trials or whether success was included as a covariate. If this information is present, it needs to be explicitly clarified. The exclusion criteria for unsuccessful trials in both tasks should also be detailed. Moreover, the decision to exclude electrodes as independent variables in the models warrants an explanation.
We appreciate the opportunity to clarify these points. In the revised manuscript, we have now explicitly reported the descriptive statistics and the results of a mixed-effects logistic model on response success in the revised manuscript (page 8, para. 1, lines 2–4; Supplementary Table S1). Participants achieved similarly high success rates in both self (M = 97%) and other trials (M = 96%; Figure S3). As shown in Table S2, success rates decreased as effort increased (b = -4.77, p < 0.001). However, no other effects reached significance (ps > 0.245). These near-ceiling success rates indicate strong task engagement and effectively rule out risk aversion as a potential confound.
Regarding model structure, we excluded unsuccessful trials from statistical analyses because they were rare and distributed equally across conditions. Given the near-ceiling performance, we did not include success rate as a covariate, as it offers limited variance.
Finally, we did not include electrodes as an independent variable because our hypotheses focused on condition effects rather than topographic differences. Following established research (e.g., Krigolson, 2018; Proudfit, 2015), we averaged RewP amplitudes across a frontocentral cluster (FC3, FCz, and FC4) and P3 amplitudes across a parietal cluster (P3, Pz, and P4), where activity is typically maximal. Averaging across these theoretically grounded clusters improves the signal-to-noise ratio and provides more reliable estimates of the underlying components. We have explicitly included this rationale in the revised manuscript, which reads, “Data were averaged across the selected electrode clusters to improve signal-to-noise ratio and reliability” (page 27, para. 1, lines 9–10).
Proudfit, G. H. (2015). The reward positivity: From basic research on reward to a biomarker for depression. Psychophysiology, 52(4), 449-459. https://doi.org/10.1111/psyp.12370
Krigolson, O. E. (2018). Event-related brain potentials and the study of reward processing: Methodological considerations. Int J Psychophysiol, 132(Pt B), 175-183. https://doi.org/10.1016/j.ijpsycho.2017.11.007
(4) Prosocial Decision Computational Modelling: The prosocial decision task largely replicates prior behavioural findings but misses the opportunity to directly test the hypotheses derived from neural data in the prosocial effort task. If the authors propose a paradoxical effect of effort on self-rewards and an inverse effect for prosocial effort, this could be formalised in a computational model. A model comparison could evaluate the proposed mechanism against alternative theories, incorporating the complex interplay of effort and reward for self and others. Furthermore, these parameters should be correlated with neural signals, adding a critical layer of evidence to the claims. As it is, the inclusion of the prosocial decision task seems irrelevant.
We thank Reviewer #3 for this thoughtful suggestion regarding the value of computational modelling. We fully agree that formalizing mechanisms is crucial, but we would like to clarify why a computational model of decision-making cannot directly capture the paradoxical after-effects observed in our neural data. The paradoxical after-effect of effort exertion we report refers to experienced utility (i.e., how prior costs modulate the hedonic consumption of a reward), whereas the decision task measures decision utility (i.e., how prospective costs and benefits are integrated to guide choice). We included the prosocial decision task to establish a behavioral baseline and replicate the well-documented phenomenon of prosocial apathy. Consistent with prior work (e.g., Lockwood et al., 2017; Lockwood et al., 2022), our data show that at the decision stage (ex-ante), effort functions as a universal cost: participants discounted rewards for both self and others, differing only quantitatively (steeper discounting for others). It is only after effort is exerted (ex-post) that the pattern reverses: effort is valued for self but remains costly for others, representing a qualitative shift. Crucially, incorporating a "paradoxical valuation" parameter (i.e., effort as a reward) into our decision model would mathematically contradict the behavioral reality. Since participants actively avoided high-effort options, a model assuming effort adds value might fail to fit the choice data. The theoretical novelty of our study lies precisely in this temporal dissociation: whereas self-benefiting effort paradoxically enhances reward valuation, other-benefiting effort induces a persistent reward devaluation.
To address the reviewer’s interest in bridging these two domains, we examined whether these distinct stages are linked at the level of individual differences. We hypothesized that an individual’s sensitivity to prospective effort cost (discounting rate K) might modulate their susceptibility to the retrospective neural after-effect. As detailed in our Responses to Reviewer #3’s point 2, we found that for self-benefiting trials, high-discounting individuals showed an effort-enhancement effect on the RewP at low reward magnitude, while for other-benefiting trials, low-discounting individuals exhibited effort-discounting effects at high reward magnitude. We sincerely thank Reviewer #3’ for this valuable suggestion, which has successfully correlated the two tasks and facilitated our understanding of the mechanisms at play.
Lockwood, P. L., Hamonet, M., Zhang, S. H., Ratnavel, A., Salmony, F. U., Husain, M., & Apps, M. A. J. (2017). Prosocial apathy for helping others when effort is required. Nat Hum Behav, 1(7), 0131. https://doi.org/10.1038/s41562-017-0131.
Lockwood, P. L., Wittmann, M. K., Nili, H., Matsumoto-Ryan, M., Abdurahman, A., Cutler, J., Husain, M., & Apps, M. A. J. (2022). Distinct neural representations for prosocial and self-benefiting effort. Curr Biol, 32(19), 4172-4185 e4177. https://doi.org/10.1016/j.cub.2022.08.010.
(5) Contradiction Between Effort Perception and Neural Results: Participants reported effort as less effortful in the prosocial condition compared to the self condition, which seems contradictory to the neural findings and the authors' interpretation. If effort has a discounting effect on rewards for others, one might expect it to feel more effortful. How do the authors reconcile these results? Additionally, the relationship between behavioural data and neural responses should be examined to clarify these inconsistencies.
This point aligns with the issues raised in Reviewer #1’s point 2. We acknowledge the apparent discrepancy between lower reported effort in the prosocial condition and the neural discounting effect. As detailed in our Responses to Reviewer #1’s point 2, we reconcile this by proposing that subjective effort ratings in this paradigm likely reflect motivational engagement (e.g., “how hard I tried” or “how willing I was”) rather than perceived task demands. Under this interpretation, the lower effort ratings for others reflect a withdrawal of engagement (consistent with prosocial apathy), which conceptually aligns with, rather than contradicts, the neural discounting effect. To validate this, we contrasted effort ratings with difficulty ratings (a more reliable index of objective demand). Our correlational analysis revealed no significant relationship between difficulty and effort ratings (r = -0.21, p = 0.196), suggesting that they capture distinct constructs. Furthermore, liking ratings were negatively correlated with difficulty ratings (r = -0.43, p = 0.011) but not with effort ratings (r = 0.32, p = 0.061), further dissociating the two measures. Crucially, as detailed in our Responses to Reviewer #1’s Recommendations point 5, our RewP effects remained significant even after controlling for individual effort ratings. This demonstrates that the neural effort-discounting effect for others is a physiological signature that operates independently of the subjective report bias.
(6) Necessary Revisions to Manuscript: If the authors address the issues above, corresponding updates to the introduction and discussion sections could strengthen the narrative and align the manuscript with the additional analyses.
We thank Reviewer #3 for the above insightful and helpful comments. We have carefully addressed these issues raised above and have updated the manuscript accordingly, including abstract, introduction, result, and discussion sections.
Recommendations for the Authors:
Reviewer #1 (Recommendations for the authors):
Major comments:
(1) The two biggest concerns I have are
- Whether the mixed-effect models are properly specified, and
- Whether the main interaction between the Recipient and effort on the reward positivity (RewP) reflects different levels of effort exertion when working for self versus others.
We thank Reviewer #1 for identifying these two critical issues. We have carefully considered these points and conducted additional analyses to address them. Below, we provide a detailed response to each concern, explaining how we have improved the model specification and ruled out alternative interpretations regarding effort exertion.
(2) On the first point, I noticed that the authors selectively excluded random effects for Effort and Magnitude when regressing RewP on Effort, Magnitude, Recipient, and Valence. This is important because the key result in the paper is a fixed effect two-way interaction between Recipient and Effort and a three-way interaction between Recipient, Effort, and Magnitude. It is not clear that these results will remain significant when Effort and Magnitude are included as random effects in the model. Thus the authors should justify their exclusion as random effects, and/or show that the results don't depend on including those random effects in the model. The same logic applies to the specification of other mixed effects models (e.g. the effect of Magnitude in the model predicting RTs).
We thank Reviewer #1 for raising this important methodological point. We fully agree that including random slopes wherever possible reduces Type 1 error rates and yields more conservative tests of fixed effects. In our analyses, we determined the random effects structure for each model using singular value decomposition (SVD). Specifically, we began with a maximal model that included by-participant random slopes for all main effects and interactions as well as a participant-level random intercept. When the model failed to converge or yielded a singular fit, we applied SVD to identify redundant dimensions (i.e., components explaining zero variance) and iteratively removed these terms until convergence was achieved. This procedure allowed us to retain the maximal converging random-effects structure while ensuring model stability. We have clarified this procedure in the revised manuscript as follows, “For each model, we fitted the maximal random-effects structure and, when the model was overparameterized, used singular value decomposition to simplify the random-effects structure until the model converged” (page 28, para. 1, lines 5–8).
Regarding the RewP model, including all variables (i.e., Recipient, Effort, Magnitude, and Valence) in the random-effects structure resulted in a boundary (singular) fit. Examination of the variance-covariance structure of the random effects revealed that the random slopes for Valence and Magnitude were perfectly negatively correlated (r = -1.00), indicating severe overparameterization. In our original submission, we removed the random slopes for Effort and Magnitude because the SVD analysis indicated redundant dimensions in the model structure.
However, we agree with the Reviewer that retaining slopes for variables involved in key interactions is crucial. Therefore, we re-evaluated the model strategy: instead of removing Effort and Magnitude, we removed the random slope for Valence (which was the primary source of the perfect correlation). This modification successfully resolved the singularity while allowing us to retain the random slopes for the critical variables (i.e., Effort and Magnitude).
Critically, this updated model yielded the same pattern of results as our original submission: the two-way interaction between Recipient and Effort and the three-way interaction between Recipient, Effort, and Magnitude remained significant (see Table S3). As expected, including the random slopes for Effort and Magnitude yielded a more conservative test of the fixed effects. While the critical three-way interaction remained significant (p = 0.019), the simple slope for the Self condition at high reward magnitude shifted slightly from significant (p = 0.041) to marginally significant (p = 0.056). However, the effect size remained largely unchanged (b = 0.42 vs. original b = 0.43), and the dissociation pattern, where self-benefiting trials show a positive trend while other-benefiting trials show a significant negative slope, remains robust and is statistically supported by the significant interaction. We have adopted this updated model in the revised manuscript and updated the relevant sections accordingly. Finally, note that we have removed the RewP table from the Supplementary Materials because the RewP model results are now presented as a figure in the main text (as suggested by Reviewer #1’s Recommendations point 3).
We have also carefully verified the random effects structures for other mixed-effects models, including the RT and Performance-P3 models in the prosocial effort task, as well as the decision time and decision choice models in the prosocial decision-making task. The updated information is detailed as follows:
Regarding the RT model, we replaced it with a more reasonable model of response speed (button presses per second), as suggested by Reviewer #1 (see our responses to Reviewer #1’s Recommendations point 4 for details).
Regarding the performance-P3 model, the random-effects structure could only support Effort, as in our original submission; thus, the results remain unchanged.
Regarding the decision time model, we have updated our results to include the quadratic effort term, as suggested by Reviewer #1 (see our responses to Reviewer #1’s Recommendations point 6 for details).
Regarding the decision choice model, we included Recipient, Effort, and Magnitude in the random-effects structure. As shown in Table S4, the results remain largely consistent with the original model, except for a newly significant interaction between effort and magnitude. Follow-up simple slopes analyses revealed that the discounted effect of effort was more pronounced at low reward magnitude (M − 1SD: b = -2.69, 95% CI = [-3.09, -2.29], p < 0.001) than at high reward magnitude (M + 1SD: b = -2.38, 95% CI = [-2.82, -1.94],p < 0.001).
In summary, we have improved the model specification following Reviewer #1’s suggestion. Crucially, the results remain qualitatively consistent with our original findings. We have updated the Results section, figures (Figures 2, 4, and 5), and OSF documents (including a new R Markdown file and an HTML output file detailing the final results) to reflect these analyses. Additionally, we have explicitly stated the method used for calculating p-values in the mixed-effects models (page 28, para. 1, lines 8–10), which was omitted in the original submission.
(3) Regarding the mixed models, it would also be good to show a graphical depiction summarizing key effects (e.g. the Recipient by Effort interaction on RewP) rather than just showing the predictions of the fitted mixed effects models.
This point is well-taken. Please see Figure S4, which visualizes the key effects and has now been included in the revised manuscript as Figure 4A.
(4) Finally, regarding the mixed effect models of RTs - given the common finding that RTs are not normally distributed, the Authors might be better off regressing 1/RT (interpreted as speed rather than latency) since 1/RT will often make distributions less asymmetric and heavy-tailed.
We thank Reviewer #1 for this helpful suggestion regarding data distribution. In our original analysis, the dependent variable was “completion time” (i.e., the latency to complete the required button presses with the 6-s window). We agree that these raw latency data exhibited characteristic non-normality (see Figure S5, Left). Based on Reviewer #1’s suggestion, we adopted “response speed” (calculated as button presses per second) as the dependent variable. As expected, this transformation substantially improved the normality of the distribution (see Figure S5, Right). We have refitted the mixed-effects model using this speed metric. Critically, the results largely replicated the patterns observed in our original model, with the exception that the main effect of reward magnitude did not reach significance in the speed model (see Table 5). Given the superior distributional properties of the speed metric, we have replaced the original latency analysis with the response speed model in the revised manuscript. We have updated the Results section (page 8, para. 1, lines 4–9) and Figures 2B–C accordingly.
(5) Regarding the level of effort exerted, there are two reasons to suspect that participants exerted less for others versus themselves. The first is that they were slower to complete the button pressing for others versus themselves. The second is that they reported paradoxically less subjective effort for others versus self (paradoxical because they also reported liking the task less for others versus self). The explanation for both may be that they exerted less effort for others versus self and this has important implications for interpreting the main effects. If they exerted less effort for others, this may partly account for the key Recipient:Effort and Recipient:Effort:Magnitude interactions in the mixed effects regression of RewP. Do either median effort durations or self-reported effort predict the magnitude of the Recipient:Effort and Recipient:Effort:Magnitude interactions (if these were included as random effects)? If so, that would provide evidence supporting this story. Alternatively, if median durations or self-reported effort were included as covariates, do these interactions still obtain? In any case, the Authors should include caveats regarding this potential explanation of the self-versus-other interactions with effort and magnitude on the RewP" (or explain why this can not explain the interactions).
We thank Reviewer #1 for raising this important interpretational issue. We acknowledge the concern that differences in physical exertion or perceived effort could potentially confound the neural findings. However, we argue that the observed RewP effects are not driven by these factors for several reasons.
First, the prosocial effort task enforced fixed effort thresholds (10%–90% of their maximum effort level) across self-benefiting and other-benefiting trials. Importantly, participants achieved ceiling-level success rates that were highly comparable between self-benefiting (97%) and other-benefiting (96%) trials, indicating that they successfully exerted the required effort across conditions.
Second, regarding the slower response speed for others (we used response speed instead of completion time, as the former is more suitable for statistical analysis; see details in Responses to Reviewer #1’s Recommendations point 4), we interpret this as a reduction in motivation rather than a reduction in the amount of effort exerted. Similarly, as detailed in our Responses to Reviewer#1’s point 2, subjective effort ratings in this paradigm appear to be influenced by demand characteristics and do not reliably track physical exertion. For instance, liking ratings were associated with difficulty (r = -0.43, p = 0.011) instead of effort (r = 0.32, p = 0.061) ratings.
To empirically rule out the possibility that these behavioral differences account for the neural effect, we followed the reviewer’s suggestion and re-ran the mixed-effects model predicting RewP amplitudes with trial-by-trial response speed and subjective effort rating included as covariates. These control analyses revealed that neither response speed (b = -0.07, p = 0.614) nor self-reported effort (b = 0.10, p = 0.186) significantly predicted RewP amplitudes (see Table S6). Most importantly, the key interactions of interest (Recipient × Effort and Recipient × Effort × Magnitude) remained significant and virtually unchanged. These findings suggest that the observed neural after-effects of prosocial effort are not driven by variations in motor execution or perceived effort.
Minor comments:
(6) In Figure 5A a quadratic effect (not a linear effect) seems fairly obvious in decision times as a function of effort level. This makes sense given that participants are close to indifference, on average, around the 50-70% effort level. I recommend fitting a model that has a quadratic predictor and not just a linear predictor when regression decision times on effort levels.
We thank Reviewer #1 for this insightful suggestion. We agree that decision times likely track decision conflict, which typically peaks near indifference points (e.g., moderate effort levels). Accordingly, we reanalyzed the decision time data using a mixed-effects model that included both linear and quadratic terms for effort. As detailed in Table S7, this analysis revealed a significant quadratic main effect of effort, which was further qualified by a significant interaction between the quadratic effort term and reward magnitude. Decomposition of this interaction (Figure S6) revealed that the quadratic effort effect was more pronounced at low reward magnitude (M − 1SD: b = -160.10, 95% CI = [-218.30, -101.90], p < 0.001) than at high reward magnitude (M + 1SD: b = -99.50, 95% CI = [-157.60, -41.40], p = 0.001). However, we found no significant interactions involving the quadratic effort term and recipient. We have updated the Results section (page 13, para. 2; page 14, para. 1) and Figures 5A–B (right panel) to reflect these findings.
(7) The distinction between the effort and decision-making tasks wasn't super clear from the main text. A sentence early on in the results section could be useful for readers' understanding.
This point is well taken. In the revised manuscript, we have clarified this distinction at the beginning of the Results section (page 6, para. 2, lines 1–10). In addition, we have explicitly indicated the corresponding task within each subsection heading in the Results:
“2.1 Investing effort for others is less motivating than for self in the prosocial effort task” (page 7)
“2.2 Effort adds reward value for self but discounts reward value for others in the prosocial effort task” (page 9)
“2.3 Reward is devalued by effort to a higher degree for others than for self in the prosocial decision-making task” (page 13)
(8) To what does "three trials" refer to on lines 143-144?
Thank you for raising this point. Participants completed three trials in which they were asked to press a button as rapidly as possible with their non-dominant pinky finger for 6000 ms. The maximum effort level was operationalized as the average button-press count across the three trials. To improve clarity, we have also provided more detailed description in the Results section, which reads: “The mean maximum effort level (i.e., the average button-press count across three 6000-ms trials; see Procedure for details) ….” (page 7, para. 1, lines 1–2).
(9) It is unclear how the authors select their time windows for ERP analyses.
We thank Reviewer #1 for this comment. Measurement parameters (i.e., time windows and channel sites) were determined based on the grand-averaged ERP waveforms and topographic maps collapsed across all conditions. This procedure is orthogonal to the conditions of interest and prevents bias in the selection of measurement windows and channels, consistent with the “orthogonal selection approach” (Luck & Gaspelin, 2017). We have clarified this point in the revised manuscript, which now reads, “Measurement parameters (time windows and channel sites) were determined from the grand-averaged ERP waveforms and topographic maps collapsed across all conditions, which was thus orthogonal to the conditions of interest (Luck & Gaspelin, 2017)” (page 27, para. 1, lines 6–9).
Luck, S., & Gaspelin, N. (2017). How to get statistically significant effects in any ERP experiment (and why you shouldn't). Psychophysiology, 54(1), 146-157.
(10) There are a few typos throughout. For example, Line 124 should read "other half benefitted...", Line 127 should read "interest at each effort level...", "following" on Line 369, and Supplemental table titles incorrectly spell the word "Results".
We thank Reviewer #1 for catching these errors. We have corrected all the specific typos noted (page 6, para. 2, lines 11 and 15; page 22, para. 3, line 2; Supplementary Table S2). Furthermore, we have conducted a thorough proofreading of the entire text and supplementary materials to ensure linguistic accuracy and consistency throughout the manuscript.
Reviewer #2 (Recommendations for the authors):
Minor comments:
(1) Lines 84-86. "The RewP ... has its neural sources in the anterior cingulate cortex (Gehring & Willoughby, 2002) and ventral striatum (Foti et al., 2011)." This is a better reference for the ACC source: https://pubmed.ncbi.nlm.nih.gov/23973408/. And perhaps remove the reference to the ventral striatum; most people would agree that activity in the ventral striatum cannot be measured with scalp EEG.
We thank Reviewer #2 for providing the updated reference, which has been cited in the revised manuscript. We agree that activity in the VS cannot be reliably measured with scalp EEG and thus have removed the reference to the VS. The revised sentence now reads, “… has its neural sources in the anterior cingulate cortex (Gehring & Willoughby, 2002; Hauser et al., 2014)” (page 4, para. 2, lines 12–13).
(2) Lines 152-153. What exactly is shown in Figure 2A? How did the authors average across subjects?
We thank Reviewer #2 for raising this issue. Figure 2A depicts the distribution of the maximum effort level, defined as the average button-press count across three 6000-ms trials completed before the prosocial effort task. In these trials, participants were instructed to press the button as rapidly as possible with their non-dominant pinky fingers. To improve clarity, we have revised the figure caption as: “(A) Distribution of the maximum effort level (i.e., the average button-press count across three 6000-ms trials) across participants” (Figure 2).
(3) Lines 160-164. "As expected (Figure 2D), participants perceived increased effort as more difficult ... and more disliking (b = -0.62, p < 0.001) when the beneficiary was others than themselves." Does this sentence describe the main effect of the beneficiary or the interaction between beneficiary and effort level, as the start of the sentence ("increased effort") suggests?
We thank Reviewer #2 for pointing out this ambiguity. The sentence describes the main effect of beneficiary rather than the interaction between beneficiary and effort level. In the revised manuscript, we have rephrased the sentence as: “They felt less effort (b = -0.32, p = 0.019) and more disliking (b = -0.62, p = 0.001) for other-benefiting trials compared to self-benefiting trials” (page 9, para. 1, lines 4–6).
(4) Lines 195-196. "..., we conducted post-hoc simple slopes analyses at -1 SD ("Low") and + SD ("High") reward magnitude." I did not understand what the authors meant with these reward magnitudes, given that the actual potential rewards were ¥0.2, ¥0.4, ¥0.6, ¥0.8, and ¥1.0.
In our analyses, the actual reward magnitudes (¥0.2, ¥0.4, ¥0.6, ¥0.8, and ¥1.0) were z-scored and entered as a continuous regressor in the mixed-effects models. Post-hoc simple slopes analyses were then conducted at ±1 SD from the mean of the z-scored reward magnitude. To clarify, we have revised the sentence as “… we conducted post-hoc simple slopes analyses at 1 standard deviation (SD) below (“Low”) and above (“High”) the mean reward magnitude” (page 11, para. 2, lines 8–9). This standard method for testing simple effects for continuous predictors is recommended by Aiken and West (1991). Aiken, L. S., West, S. G., & Reno, R. R. (1991). Multiple regression: Testing and interpreting interactions. Sage.
(5) Lines 253 and 275. I would not call this a computational model. The authors fit a curve to data, there is no model of the computations involved.
This point is well taken. We have replaced “computational model” with “discounting” (Figure 5) and “parabolic discounting model” (page 15, para. 1, line 15).
(6) Line 710. Figure S1 does not show topographic maps of the P3, as the figure caption suggests.
We thank Reviewer #2 for identifying this oversight. We have now included topographic maps of the P3 in Figure S1.
(7) Please check language in lines 33 (effect between), 38 (shape), 49 (highest cost form?), 74 (tunning), 90 (omit following), 127 (interest on at each effort level), 135 (press buttons >> rapidly press a button?), 142 (motivated), 219 (should low be high?), 265-266 (missing word), 275 (confirmed by following), 292 (an action can be effortful, a feeling cannot), 315 (when it comes into), 330-331 (data is plural; the aftereffect of prosocial effect), 387 (interest on at each effort level), 405 (should quickly be often?).
We thank Reviewer #2 for the careful review and feedback about these language issues. We have revised all the phrasing you identified. The corrections are as follows:
Line 33: “effect between” has been changed to “effects for” (page 2, para. 1, line 6).
Line 38: “shape” has been updated to “shapes” (page 2, para. 1, line 13).
Line 49: “highest cost form?” has been revised to “the most common cost type” (page 3, para. 1, lines 7–8).
Line 74: “tunning” has been corrected to “tuning” (page 4, para. 2, line 1).
Line 90: omit following. Done (page 5, para. 1, line 2).
Line 127: “interest on at each effort level” has been corrected to “liking for each effort level” (page 6, para. 2, line 15).
Line 135: “press buttons” has been updated to “rapidly press a button” (the caption of Figure 1).
Line 142: “motivated” has been revised to “motivating” (page 7).
Line 219: should low be high? Yes, we have corrected this (the caption of Figure 4).
Lines 265–266: The missing word “with” has been inserted (page 15, para. 1, line 2).
Line 275: “confirmed by following” has been revised as “corroborated by a parabolic …” (page 15, para. 1, line 15).
Line 292: an action can be effortful, a feeling cannot. We have changed the word “effortful” to “effort” (page 18, para. 2, line 3).
Line 315: “when it comes into” has been revised to “when it came to” (page 19, para. 1, line 10).
Lines 330–331: These two expressions have been revised to “our data establish …” and “the after-effect of prosocial effort” (page 20, para. 1, lines 2–3).
Line 387: “interest on at each effort level” has been corrected to “interest at each effort level” (page 23, para. 2, line 5).
Line 405: should quickly be often? We agree that “quickly” might imply latency or speed of a single press, whereas the task required maximizing the frequency of presses within the time window. To capture this meaning accurately, we have revised the phrase to “pressed a button as rapidly as possible” (implying repetition rate) in the revised manuscript (page 24, para. 2, lines 3–4).
-
eLife Assessment
The findings are considered valuable and have theoretical implications for the interdisciplinary field of value-based social decision-making. Support for the main claims is incomplete and these should be supported by further analyses.
-
Reviewer #1 (Public review):
Summary:
The authors test the hypotheses, using an effort-exertion and an effort-based decision-making task, while recording brain dynamics with EEG, that the brain processes reward outcomes for effort differentially when they earned for themselves versus others.
Strengths:
The strengths of this experiment include what appears to be a novel finding of opposite signed effects of effort on the processing of reward outcomes when the recipient is self versus others. Also, the experiment is well-designed, the study seems sufficiently powered, and the data and code are publicly available.
Weaknesses:
Inferences rely heavily on the results of mixed effects models which may or may not be properly specified and are not supported by complementary analyses. Also, not all results hang together in a sensible way. For …
Reviewer #1 (Public review):
Summary:
The authors test the hypotheses, using an effort-exertion and an effort-based decision-making task, while recording brain dynamics with EEG, that the brain processes reward outcomes for effort differentially when they earned for themselves versus others.
Strengths:
The strengths of this experiment include what appears to be a novel finding of opposite signed effects of effort on the processing of reward outcomes when the recipient is self versus others. Also, the experiment is well-designed, the study seems sufficiently powered, and the data and code are publicly available.
Weaknesses:
Inferences rely heavily on the results of mixed effects models which may or may not be properly specified and are not supported by complementary analyses. Also, not all results hang together in a sensible way. For example, participants report feeling less subjective effort, but also more disliking of tasks when they were earning rewards for others versus self. Given that participants took longer to complete tasks when earning effort for others, it is conceivable that participants might have been working less hard for others versus themselves, and this may complicate the interpretation of results.
-
Reviewer #2 (Public review):
Summary:
Measurements of the reward positivity, an electrophysiological component elicited during reward evaluation, have previously been used to understand how self-benefitting effort expenditure influences the processing of rewards. The present study is the first to complement those measurements with electrophysiological reward after-effects of effort expenditure during prosocial acts. The results provide solid evidence that effort adds reward value when the recipient of the reward is the self but discounts reward value when the beneficiary is another individual.
Strengths:
An important strength of the study is that the amount of effort, the prospective reward, the recipient of the reward, and whether the reward was actually gained or not were parametrically and orthogonally varied. In addition, the …
Reviewer #2 (Public review):
Summary:
Measurements of the reward positivity, an electrophysiological component elicited during reward evaluation, have previously been used to understand how self-benefitting effort expenditure influences the processing of rewards. The present study is the first to complement those measurements with electrophysiological reward after-effects of effort expenditure during prosocial acts. The results provide solid evidence that effort adds reward value when the recipient of the reward is the self but discounts reward value when the beneficiary is another individual.
Strengths:
An important strength of the study is that the amount of effort, the prospective reward, the recipient of the reward, and whether the reward was actually gained or not were parametrically and orthogonally varied. In addition, the researchers examined whether the pattern of results generalized to decisions about future efforts. The sample size (N=40) and mixed-effects regression models are also appropriate for addressing the key research questions. Those conclusions are plausible and adequately supported by statistical analyses.
Weaknesses:
Although the obtained results are highly plausible, I am concerned whether the reward positivity (RewP) and P3 were adequately measured. The RewP and P3 were defined as the average voltage values in the time intervals 300-400 ms and 300-440 ms after feedback onset, respectively. So they largely overlapped in time. Although the RewP measure was based on frontocentral electrodes (FC3, FCz, and FC4) and the P3 on posterior electrodes (P3, Pz, and P4), the scalp topographies in Figure 3 show that the RewP effects were larger at the posterior electrodes used for the P3 than at frontocentral electrodes. So there is a concern that the RewP and P3 were not independently measured. This type of problem can often be resolved using a spatiotemporal principal component analysis. My faith in the conclusions drawn would be further strengthened if the researchers extracted separate principal components for the RewP and P3 and performed their statistical analyses on the corresponding factor scores.
-
Reviewer #3 (Public review):
This study investigates how effort influences reward evaluation during prosocial behaviour using EEG and experimental tasks manipulating effort and rewards for self and others. Results reveal a dissociable effect: for self-benefitting effort, rewards are evaluated more positively as effort increases, while for other-benefitting effort, rewards are evaluated less positively with higher effort. This dissociation, driven by reward system activation and independent of performance, provides new insights into the neural mechanisms of effort and reward in prosocial contexts.
This work makes a valuable contribution to the prosocial behaviour literature by addressing areas that previous research has largely overlooked. It highlights the paradoxical effect of effort on reward evaluation and opens new avenues for …
Reviewer #3 (Public review):
This study investigates how effort influences reward evaluation during prosocial behaviour using EEG and experimental tasks manipulating effort and rewards for self and others. Results reveal a dissociable effect: for self-benefitting effort, rewards are evaluated more positively as effort increases, while for other-benefitting effort, rewards are evaluated less positively with higher effort. This dissociation, driven by reward system activation and independent of performance, provides new insights into the neural mechanisms of effort and reward in prosocial contexts.
This work makes a valuable contribution to the prosocial behaviour literature by addressing areas that previous research has largely overlooked. It highlights the paradoxical effect of effort on reward evaluation and opens new avenues for investigating the mechanisms underlying this phenomenon. The study employs well-established tasks with robust replication in the literature and innovatively incorporates ERPs to examine effort-based prosocial decision-making - an area insufficiently explored in prior work. Moreover, the analyses are rigorous and grounded in established methodologies, further enhancing the study's credibility. These elements collectively underscore the study's significance in advancing our understanding of effort-based decision-making.
Despite these contributions, there are several gaps in the analysis that leave the conclusions incomplete and warrant further investigation. These issues can be summarized as follows:
(1) Incomplete EEG Reporting: The methods indicate that EEG activity was recorded for both tasks; however, the manuscript reports EEG results only for the first task, omitting the decision-making task. If the authors claim a paradoxical effect of effort on self versus other rewards, as revealed by the RewP component, this should also be confirmed with results from the decision-making task. Omitting these findings weakens the overall argument.
(2) Neural and Behavioural Integration: The neural results should be contrasted with behavioural data both within and between tasks. Specifically, the manuscript could examine whether neural responses predict performance within each task and whether neural and behavioural signals correlate across tasks. This integration would provide a more comprehensive understanding of the mechanisms at play.
(3) Success Rate and Model Structure: The manuscript does not clearly report the success rate in the prosocial effort task. If success rates are low, risk aversion could confound the results. Additionally, it is unclear whether the models accounted for successful versus unsuccessful trials or whether success was included as a covariate. If this information is present, it needs to be explicitly clarified. The exclusion criteria for unsuccessful trials in both tasks should also be detailed. Moreover, the decision to exclude electrodes as independent variables in the models warrants an explanation.
(4) Prosocial Decision Computational Modelling: The prosocial decision task largely replicates prior behavioural findings but misses the opportunity to directly test the hypotheses derived from neural data in the prosocial effort task. If the authors propose a paradoxical effect of effort on self-rewards and an inverse effect for prosocial effort, this could be formalised in a computational model. A model comparison could evaluate the proposed mechanism against alternative theories, incorporating the complex interplay of effort and reward for self and others. Furthermore, these parameters should be correlated with neural signals, adding a critical layer of evidence to the claims. As it is, the inclusion of the prosocial decision task seems irrelevant.
(5) Contradiction Between Effort Perception and Neural Results: Participants reported effort as less effortful in the prosocial condition compared to the self condition, which seems contradictory to the neural findings and the authors' interpretation. If effort has a discounting effect on rewards for others, one might expect it to feel more effortful. How do the authors reconcile these results? Additionally, the relationship between behavioural data and neural responses should be examined to clarify these inconsistencies.
Necessary Revisions to Manuscript: If the authors address the issues above, corresponding updates to the introduction and discussion sections could strengthen the narrative and align the manuscript with the additional analyses.
-
-
