Selected work
| Year | Work | Source |
|---|---|---|
| 2026 | AI as amplifier in sensory and consumer science (accepted) | paper 2026 |
| 2026 | Created Aigora’s AI course, The AI Amplified Scientist, which I teach with Vanessa Rios de Souza | course page |
| 2026 | Does more model “thinking” help AI-supported sensory analysis? (in review) | materials |
| 2026 | Two preprints on horoconvex domains, with X. Dai, X. H. Nguyen, and G. Wei | arXiv 2606.12745 · arXiv 2607.27120 |
| 2025 | Created Theus, Aigora’s research memory platform | Theus page |
| 2025 | Returned to Aigora | |
| 2023 | Data Science for Sensory and Consumer Scientists, a book with Thierry Worch, Julien Delarue, and Vanessa Rios de Souza | book 2023 |
| 2023 | Pre-seed round for NeoSwap AI, the venture-backed AI startup I co-founded and ran as CEO | funding · Tulle 2024 |
| 2020 | Consumer surveys on Alexa, before ChatGPT | AigoraCast 2020 |
| 2019 | Founded Aigora | about |
| 2013 | Food Quality and Preference Award for Contributions by a Researcher of the Future | award page |
| 2012 | TURF by integer linear programming; compact letter displays by clique covering | paper 2012 · paper 2012 |
| 2011 | The tetrad test has, in theory, more power than the triangle test; this “sparked a lot of activity concerning the tetrad” (Michael O’Mahony) | paper 2011 · O’Mahony 2013 |
| 2010 | The universal cover of a noncompact limit space, with Guofang Wei | paper 2010 |
| 2007 | SPEED, a model of automatic categorization, with F. Gregory Ashby | paper 2007 |
| 2006 | The universal cover of a compact limit space, with Guofang Wei | paper 2006 |
| 1998 | First paper: Thurstonian models for the tetrad method, with Daniel M. Ennis | paper 1998 |
Publications
2020s
2026
- Sensing the path forward: AI as amplifier in sensory and consumer science (accepted)doi:10.1016/j.foodqual.2026.106105
- Smarter fragrance innovation: Using machine learning to predict and optimize consumer likingdoi:10.1016/j.foodqual.2026.106032 (corrigendum)
- Centralized first-price clearing for budget-constrained two-sided multi-item marketsdoi:10.23919/fruct70069.2026.11506544
- Log-concavity and level-set horoconvexity of the first eigenfunction on horoconvex domains in the hyperbolic planearXiv:2607.27120
- A large-diameter fundamental-gap lower bound for horoconvex domainsarXiv:2606.12745
2024
- Using Web3 technologies to represent personalized consumer taste preferences in whiskiesdoi:10.1016/j.foodqual.2024.105201
2023
- Data science for sensory and consumer scientistsdoi:10.1201/9781003028611
- Understanding the effects of smart-speaker-based surveys on panelist experience in immersive consumer testingdoi:10.3390/foods12132537
- Voice-activated technology in sensory and consumer research: A new frontierdoi:10.1016/b978-0-323-95225-5.00017-1
2022
- Sensory interaction effects between capsicum heat and seasoning ingredients applied to unsalted potato chipsdoi:10.1016/j.foodqual.2022.104682
2021
- Emotional profiles elicited from orthonasal and retronasal perceptions of food (fruit) and non-food (floral) aromasdoi:10.1002/ffj.3655
2020
- Comparison of preference mapping with projective mapping for characterizing consumer perception of brewed black coffeesdoi:10.1111/joss.12563
2010s
2015
- A Thurstonian comparison of the tetrad and degree of difference testsdoi:10.1016/j.foodqual.2014.05.004
2014
- Sensory difference tests as measurement instruments: A review of recent advancesdoi:10.1111/joss.12086
- Precision of measurement in tetrad testingdoi:10.1016/j.foodqual.2013.05.003
- Paired preference data with a no-preference option: Statistical tests for comparison with placebo datadoi:10.1016/j.foodqual.2013.06.005
- Transitioning from proportion of discriminators to a more meaningful measure of sensory differencedoi:10.1016/j.foodqual.2013.04.007
2013
- A Thurstonian ranking model with rank-induced dependenciesdoi:10.1007/s00357-013-9125-8
- A Thurstonian analysis of the two-out-of-five testdoi:10.1111/joss.12044
- Analysis and Thurstonian scaling of applicability scoresdoi:10.1111/joss.12034
- Importance of correct instructions in the tetrad testdoi:10.1111/joss.12049
- Investigating the single ideal assumption using Ideal Profile Methoddoi:10.1016/j.foodqual.2013.01.005
- Mapping hedonic data: A process perspectivedoi:10.1111/joss.12048
- Reconsidering the specified tetrad testdoi:10.1111/joss.12060
- The year of the tetrad testdoi:10.1111/joss.12047
2012
- A comparison of adaptive choice-based conjoint and choice-based conjoint to determine key choice attributes of sour cream with limited sample sizedoi:10.1111/joss.12009
- A comparison of three commonly used methods for treating no preference votesdoi:10.1111/j.1745-459x.2012.00374.x
- A group level validation of the supercombinatorality property: Finding high-quality ingredient combinations using pairwise informationdoi:10.1016/j.foodqual.2012.01.003
- A large-scale experimental comparison of the tetrad and triangle tests in childrendoi:10.1111/j.1745-459x.2012.00385.x
- Accounting for no difference/preference responses or ties in choice experimentsdoi:10.1016/j.foodqual.2011.06.006
- eTURF: A competitive TURF algorithm for large datasetsdoi:10.1016/j.foodqual.2011.06.004
- Guiding the switch from triangle testing to tetrad testingdoi:10.1111/j.1745-459x.2012.00386.x
- Justifying count-based comparisonsdoi:10.1111/j.1745-459x.2012.00375.x
- Assignment-minimum clique coveringsdoi:10.1145/2133803.2275596
2011
- Confidence bounds for multiplicative comparisonsdoi:10.1080/03610921003646398
- The power of sensory discrimination methods revisiteddoi:10.1111/j.1745-459x.2011.00353.x
- Validating a graph theoretic screening approach to food item combinationsdoi:10.1111/j.1745-459x.2011.00348.x
2010
- Describing the universal cover of a noncompact limitdoi:10.2140/gt.2010.14.2479
- Equivalence hypothesis testingdoi:10.1016/j.foodqual.2009.06.005
- Equivalence hypothesis testing: Reply to Bidoi:10.1016/j.foodqual.2009.12.004
1998-2009
2009
- Hypothesis testing for equivalence defined on symmetric open intervalsdoi:10.1080/03610920802460787
2008
- Confidence bounds for positive ratios of normal random variablesdoi:10.1080/03610920701653201
2007
- A neurobiological theory of automaticity in perceptual categorizationdoi:10.1037/0033-295x.114.3.632
2006
- Describing the universal cover of a compact limitdoi:10.1016/j.difgeo.2006.05.008
- The role of the basal ganglia in category learningdoi:10.1016/s0079-7421(06)46001-1
2000
- Replicated difference and preference tests: How to account for inter-trial variationdoi:10.1016/s0950-3293(99)00032-4
1998
- Thurstonian models for variants of the method of tetradsdoi:10.1111/j.2044-8317.1998.tb00677.x
Open source
| Project | Description | Links |
|---|---|---|
| sensR | Thurstonian models for sensory discrimination tests, in R. Written by Rune Haubo Bojesen Christensen and Per Bruun Brockhoff; I am the CRAN maintainer and a listed author. | CRAN · GitHub |
| sensPy | A Python port of sensR | PyPI · GitHub |
| turfLP | TURF analysis by integer linear programming, in R, Python, and JavaScript | CRAN · app |
| cld-reducer | Reduces compact letter displays, in R, Python, and JavaScript | CRAN · demo |
| Elves | An Agent Skill that hands planned development work to a separate coding agent | guide |
| Lantern | A Herdr plugin that shows which coding agents need you | guide |
