Advancing preference testing in humans and animals

Area of Science:

• Behavioral science
• Animal welfare
• Quantitative psychology

Background:

• Preference testing ranks options but doesn't show preference strength.
• Simultaneous presentation of options causes mutual influence, increasing with more options.
• Existing methods fail to reveal the relative strengths of preference or worth value.

Purpose of the Study:

• To develop and validate a method for scaling preferences using multiple binary comparisons.
• To quantitatively assess the relative value of options presented to animals.
• To optimize preference testing for applications like animal welfare assessment.

Main Methods:

• Utilized multiple binary choice tests to scale preferences among many options.
• Analyzed human image preference data with known valence scores for validation.
• Developed an approach to incorporate additional option choices into existing rank orders via simulations.
• Introduced consensus error and intransitivity ratio for assessing scaled ranking confidence.

Main Results:

• Demonstrated that multiple binary comparisons can scale preferences, revealing worth value.
• Successfully applied the approach to assess valence in mice and rhesus macaques.
• Showed that simulations can reduce the number of animal experiments needed.
• Developed quality measures to assess and improve scaled ranking confidence.

Conclusions:

• The novel approach enables efficient and quantitative assessment of relative option values in animals.
• This method optimizes preference testing, particularly for welfare assessment.
• The developed software, an R package named "simsalRbim", is publicly available.