Impulsivity as a trait in dogs

Jeff Stevens

University of Nebraska-Lincoln


@JeffreyRStevens@fediscience.org
@JeffStevensADML

Impulsivity

Photo by Tamas Pap on Unsplash

Impulsivity is a critical component of dog (Canis familiaris) behavior that owners often want to curtail. Though studies of dog impulsivity have examined their inability to wait and to inhibit inappropriate behaviors, it is not clear whether impulsivity is a behavioral trait with consistent characteristics within individuals and across contexts. We conducted a systematic review and meta-analysis to investigate whether impulsivity exists as a behavioral trait in domestic dogs.

Is impulsivity a behavioral trait?

Photos by Gabriel Meinert on Unsplash and Guillaume de Germain on Unsplash

Is impulsivity a behavioral trait?

Systematic review and meta-analysis

1. Correlations across impulsivity behavioral tasks
2. Relationship between task performance and owner perceptions

Methods

Photo by Glenn Carstens-Peters on Unsplash

Systematic review

• Searched 9,900 records from Scopus, Web of Science, and PsychINFO for impulsivity terms
• 13 articles conducted correlations between impulsivity tasks and/or between the Dog Impulsivity Assessment Scale (DIAS) and behavioral tasks
• Extracted correlation coefficients for all pairs of tasks or task/DIAS scores

Tasks

• A-not-B Barrier
• A-not-B Cup
• Box
• Buzzer
• Cylinder
• Delay Discounting
• Detour Fence

• Leash
• Middle Cup
• Sit-Stay
• Social Inhibition
• Spatial Impulsivity
• Wait-for-Treat

Tasks

Systematic review

Correlations across impulsivity tasks

3 of 31 task pairs correlated

Task and owner perceptions

5 of 15 tasks correlated with DIAS

Meta-analysis

A-not-B Cup & Cylinder

No study correlations or overall effect

Spatial impulsivity

Two of six study correlations but no overall effect

Summary

• No evidence of widespread or consistent correlations among impulsivity tasks
• Minimal evidence of owner predictions of impulsivity

Open questions

• Is impulsivity not at trait?
• Is it multiple traits?
• Are our measures reliable?
• Are our measures valid?
• Are our sample sizes large enough?