Date of Award


Degree Name

Master of Science


Behavior Analysis and Therapy

First Advisor

Hurtado-Parrado, Camilo


Decision making is heavily influenced by the environment around us. Berry et al. (2014, 2015, 2019) showed that viewing images of natural environments during the delay discounting task resulted in lower impulsive choice, as compared to viewing images of built environments or geometric figures. Berry et al. proposed that attentional factors could explain this effect, however, recent attempts to reproduce Berry et al.’s findings in a different laboratory have been unsuccessful (Johnson 2017, 2018, 2019). The present study tested if manipulating the participants’ observing responses towards different types of images (natural, built, and no images) modulates the effect reported by Berry et al. Eighty-seven college students were exposed to a matching-to-sample task aimed at increasing observation responses to the images (attentional manipulation) throughout the same delay discounting task implemented by Berry et al. (2014). It was expected that increasing the participants interaction with the images via the matching task would increase the magnitude of the effect reported by Berry et al. (2015); namely, further reduction of impulsive responses after being exposed to images of natural environments and increase of impulsive choice when exposed to built environments. Results indicated that participants who engaged in matching images of built environments had a higher rate of discounting than the group that replicated Berry et al. (2014). Matching images of natural environments did not seem to reduce impulsive choice, as predicted based on Berry et al.’s findings. Furthermore, none of the groups of the present study reproduced the rates of delay discounting originally reported by Berry et al. Participants in Berry et al., (2014) overall discounted less steeply when compared to participants in the present study. This finding resembles the results reported by Johnson et al. (2017, 2018, 2019) in their replica attempts. Lastly, participants’ self-reports regarding time spent in natural and built environments did not correlate with rate of discounting. Future research should use the number of correct responses as a measure of procedural integrity. Also, it is possible that perhaps participants failed to match more often in the built condition, and such aversive condition could have increased the impulsive choice for that group (Flora et al. 1992, 2003).



This thesis is Open Access and may be downloaded by anyone.