Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

High Expectations: An Observational Study of Programming and Cannabis Intoxication

Wenxin He, Manasvi Parikh, Westley Weimer, Madeline Endres

TL;DR

This study uniquely quantifies the effects of ecologically valid cannabis intoxication on programming performance using a rigorous within-subject design (n=74) across sober and cannabis sessions. It finds a small-to-medium impairment in code correctness and increased time to complete tasks, with no evidence that cannabis enhances divergent or creative coding strategies. Importantly, programmers' self-assessment of performance under intoxication correlates with actual performance, suggesting limited value for broad anti-cannabis policies and highlighting the need for nuanced, evidence-based guidance for developers and organizations. The work contributes replication data and a transparent methodology to inform policy, workplace decisions, and future research on psychoactive substances in software engineering.

Abstract

Anecdotal evidence of cannabis use by professional programmers abounds. Recent studies have found that some professionals regularly use cannabis while programming even for work-related tasks. However, accounts of the impacts of cannabis on programming vary widely and are often contradictory. For example, some programmers claim that it impairs their ability to generate correct solutions while others claim it enhances creativity and focus. There remains a need for an empirical understanding of the true impacts of cannabis on programming. This paper presents the first controlled observational study of the effects of cannabis on programming ability. Based on a within-subjects design with over 70 participants, we find that at ecologically valid dosages, cannabis significantly impairs programming performance. Programs implemented while high contain more bugs and take longer to write (p < 0.05), a small to medium effect (0.22 <= d <= 0.44). We also did not find any evidence that high programmers generate more divergent solutions. However, programmers can accurately assess differences in their programming performance (r = 0.59), even when under the influence of cannabis. We hope that this research will facilitate evidence-based policies and help developers make informed decisions regarding cannabis use while programming.

High Expectations: An Observational Study of Programming and Cannabis Intoxication

TL;DR

This study uniquely quantifies the effects of ecologically valid cannabis intoxication on programming performance using a rigorous within-subject design (n=74) across sober and cannabis sessions. It finds a small-to-medium impairment in code correctness and increased time to complete tasks, with no evidence that cannabis enhances divergent or creative coding strategies. Importantly, programmers' self-assessment of performance under intoxication correlates with actual performance, suggesting limited value for broad anti-cannabis policies and highlighting the need for nuanced, evidence-based guidance for developers and organizations. The work contributes replication data and a transparent methodology to inform policy, workplace decisions, and future research on psychoactive substances in software engineering.

Abstract

Anecdotal evidence of cannabis use by professional programmers abounds. Recent studies have found that some professionals regularly use cannabis while programming even for work-related tasks. However, accounts of the impacts of cannabis on programming vary widely and are often contradictory. For example, some programmers claim that it impairs their ability to generate correct solutions while others claim it enhances creativity and focus. There remains a need for an empirical understanding of the true impacts of cannabis on programming. This paper presents the first controlled observational study of the effects of cannabis on programming ability. Based on a within-subjects design with over 70 participants, we find that at ecologically valid dosages, cannabis significantly impairs programming performance. Programs implemented while high contain more bugs and take longer to write (p < 0.05), a small to medium effect (0.22 <= d <= 0.44). We also did not find any evidence that high programmers generate more divergent solutions. However, programmers can accurately assess differences in their programming performance (r = 0.59), even when under the influence of cannabis. We hope that this research will facilitate evidence-based policies and help developers make informed decisions regarding cannabis use while programming.
Paper Structure (25 sections, 5 figures, 2 tables)

This paper contains 25 sections, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. Experimental Setup and Design
  4. Study Design
  5. Surveys and Stimuli: Content and Metrics
  6. Demographics.
  7. Short Programming Problem Stimuli
  8. "Programming Interview" Style Stimuli
  9. Participant Overview
  10. Recruitment Process
  11. Population Contextualization
  12. Research Questions and Analysis
  13. Primary Analysis: Pre-registered Questions
  14. Exploratory Analysis
  15. Statistical Methods
  16. ...and 10 more sections

Figures (5)

  • Figure 1: Example short programming stimuli, adapted from the program comprehension literature.
  • Figure 2: Example "interview-style" programming stimulus, presented in the study platform (shortened for space).
  • Figure 3: Indicative example comparing code produced while high vs. sober by the same participant for the same problem. The intoxicated code is more complicated and contains a bug.
  • Figure 4: Histograms with typed characters (dark blue with lines) and deletions (orange) over time for the same participant while sober and high. The high condition features longer pauses and more deletions. The participant also had a higher maximum typing speed sober (120 keystrokes in 30 seconds vs. only 65 while high). This participant finished both problems early sober (noted by the green box) but ran out of time when high.
  • Figure 5: Self-reported subjective programming performance in cannabis-using sessions compared to sober sessions. Most participants report perceiving decreased performance when high (63%) compared to only 17% who perceived improvement.