I wish to thank the authors for their careful consideration of the comments that were raised on the previous draft of the manuscript. I have no further comments on this revision and look forward to learning the results of this interesting research.

The authors have fully addressed all the comments made in the previous round of reviews, adding clarifications to their sample size estimation and analysis plan, and a rationale for the inclusion of memory tasks, dietary restraint and hunger as exploratory measures. 

This adds to an already high quality manuscript which I recommend for acceptance, and will look forward to reading the Stage 2 report.

Alexander MacLellan

The manuscript addresses an important and relevant question regarding GNG and AAT procedures as implemented in food-related research (and other applied domains). The introduction offers a comprehensive theoretical and empirical background, effectively leading to the study's aims and hypotheses. The methods are clearly described, ensuring reproducibility, and the analyses are detailed and well-explained. I particularly also want to emphasize that I appreciate the authors' dedication to conducting rigorous and reproducible research.

Given the manuscript's high quality, I recommend its publication with only minor revisions, as explained below.

- The authors report power analyses and ultimately decide to include 100 participants per condition (N = 200 in total). In the next paragraph detailing the sample, the authors however state that they planned to recruit 160 participants (who were at least 18 years old). So there appears to be an inconsistency here that needs clarification.

- The method also describes a memory task which is not previously mentioned in the study aims/hypotheses nor in the statistical analyses section. Similarly measures of dietary restraint and hunger are collected but not further elaborated. I presume that these would be part of the exploratory analyses? Perhaps it would be good if the authors could also provide a brief rationale for inclusion of these measures.

I am confident that the research findings will contribute to current developments in this research area both from a methodological and theoretical perspective. I look forward to reading the manuscript again once the data have been collected and analyzed.

This paper aims to investigate whether Go/NoGo cues vs Approach/Avoidance cues influence behaviour when selecting appetitive food. The question is an interesting and valid question, and the authors have a well detailed, and considered study planned. There is a clear logic and rationale, with clear implications for cognitive training programmes for unhealthy eating. The manuscript in its current state is well written, providing sufficient detail to allow for replication, and meets the majority of the issues to consider at Stage 1 of a PCI RR. My comments are as follows:

Main comments:

-          The sampling plan is very well described with an extended justification of their first sample size estimate, however the decision to recruit 100 participants per group when they state 58 in each is sufficient. Might it not be better to continue recruit until that sample size is reached (after exclusions etc.) rather than over-recruit?

-          The authors then state they will recruit 160 participants in the ‘Participants’ section (page 13), which is inconsistent with their sample size estimation.

Minor comments:

-          As a minor comment, but in the interest of transparency, it would be a positive to see a statement added at this stage as to the level of bias control achieved in this study (e.g. Level 6), as well as statements on data and code availability. 

-          The first specified analysis, ratings before the training, appears to be frequentist, but no details of frequentist assessment criteria are given.