Do dogs follow Weber’s Law?

The role of ratio and difference in quantity preference

Jeff Stevens, Hunter DeBoer, Hannah Fitzpatrick,
Anwyn Gatesy-Davis, & London Wolff

Canine Cognition and Human Interaction Lab
University of Nebraska-Lincoln
dogcog.unl.edu

Canine quantities

Photo: Mathew Coulton

Food preference task

Ward & Smuts 2007

What drives preference?

Weber’s Law

Detecting differences in stimuli depends on relative magnitude

Numerical ratio

Key signature of Weber’s law

Ratio = small / large

Small Large Ratio
1 2 0.50
2 4 0.50
1 3 0.33
2 6 0.33
Small Large Ratio
1 2 0.50
2 4 0.50
1 3 0.33
2 6 0.33

Numerical difference/distance/disparity

Difference = large − small

Small Large Difference
1 2 1
2 4 2
1 3 2
2 6 4
Small Large Difference
1 2 1
2 4 2
1 3 2
2 6 4

Previous work on difference and ratio

Study Ratio effects Difference effects
Ward & Smuts 2007 Yes Yes
Baker et al. 2012 Yes Not tested
Miletto Petrazzini & Wynn 2016 Yes Not tested
Aulet et al. 2019 Yes Not tested
Rivas-Blanco et al. 2020 Yes Not tested

Difference and ratio are confounded

Research question

What are the relative contributions of difference and ratio in determining dog quantity preferences?

Methods

Subjects

  • Recruited 13 dogs from local dog daycare facility
    • Six dogs were excluded due to disinterest or inability to complete sessions
  • Seven dogs completed all sessions
    • 2:5 male:female (all spayed/neutered)
    • Two goldendoodles, one golden retriever, one collie, three mixes
    • 1-3 years old
    • 12-78 lbs

Set-up

Procedures

  • Sessions consisted of warm-up trials, experimental trials, and jackpot trials
  • Ten sessions per subject

Numerical pairs

Small Large Difference Ratio
1 3 2 0.33
2 6 4 0.33
3 9 6 0.33
1 2 1 0.50
2 4 2 0.50
4 8 4 0.50
2 3 1 0.67
4 6 2 0.67
8 12 4 0.67
Small Large Difference Ratio
1 3 2 0.33
2 6 4 0.33
3 9 6 0.33
1 2 1 0.50
2 4 2 0.50
4 8 4 0.50
2 3 1 0.67
4 6 2 0.67
8 12 4 0.67

Results

Difference or ratio

Difference and ratio

Model BF
choice ~ (1 | dog_id)
choice ~ diff + (1 | dog_id) 2,531.3
choice ~ ratio + (1 | dog_id) 15.9
choice ~ diff + ratio + (1 | dog_id) 481.4
choice ~ diff * ratio + (1 | dog_id) 19.8
Model BF
choice ~ (1 | dog_id)
choice ~ diff + (1 | dog_id) 2,531.3
choice ~ ratio + (1 | dog_id) 15.9
choice ~ diff + ratio + (1 | dog_id) 481.4
choice ~ diff * ratio + (1 | dog_id) 19.8

Conclusions

Summary

  • Both difference and ratio predict numerical preference when analyzed separately

  • But only difference predicts when analyzed together

Previous preference work

Ward & Smuts 2007

Previous discrimination work

Rivas-Blanco et al. 2020

Implications

  • Revisit ubiquity of Weber’s Law

  • Explore other properties of Weber’s Law

  • Include difference in models and design studies to test difference effects

  • Consider differences between discrimination and preference tasks

Thank you!

Hunter DeBoer Hannah Fitzpatrick

Hunter
DeBoer Hannah
Fitzpatrick Anwyn
Gatesy-Davis London
Wolff

jstevens5@unl.edu dogcog.unl.edu @unl_cchil

Difference and ratio

Discrimination vs. preference

Cognitive Phenotypes

Mendelson et al. 2016

Discrimination vs. preference

Discrimination Preference
Stimuli Arbitrary Food rewards
Rewards All or none Presented amounts
Contingencies Correct response Chosen amounts

jeffreyrstevens.quarto.pub/csc_2024