Response Training Interventions: From Food to Substances

Extending gamified cognitive training to alcohol, tobacco, and cannabis, with a focus on real-world efficacy and public health impact.

Background: can motor training reshape the value of substance cues?

Our lab has accumulated robust evidence that Go/NoGo and approach-avoidance training can reduce the motivational pull of food cues, with preregistered trials demonstrating devaluation effects and measurable consumption reductions for sugary beverages (Najberg et al., 2021, 2023; Tapparel et al., in press). A natural and important question is whether these effects generalise to substance use, where cue-induced craving plays an equally central, and often more clinically consequential, role in driving consumption and relapse.

The theoretical basis for generalisation is strong. The incentive salience mechanisms that make food cues compulsive operate through the same dopaminergic reward circuits implicated in alcohol, tobacco, and cannabis use. If motor-based training can rewrite the motivational value of food cues by disrupting automatic approach tendencies and inducing affective retagging, the same principles should apply to substance cues. Yet the translation is not trivial: substance use populations differ from healthy food consumers in craving intensity, dependence severity, comorbidity profiles, and the regulatory and clinical context in which interventions must be delivered. Establishing whether and how response training works in these populations requires dedicated trials designed for ecological validity and real-world applicability.

Our programme

We are extending our gamified training platform to three substance domains: alcohol, cannabis, and tobacco. In each case, the same core paradigm (app-based Go/NoGo and approach-avoidance training with adaptive difficulty and gamification) is adapted with domain-specific cue sets, population-appropriate recruitment and outcome measures, and clinical partnerships that anchor the research in real-world care pathways.

A defining feature of this programme is its emphasis on effectiveness over efficacy. While our food research established the paradigm under tightly controlled laboratory conditions, the substance programme is designed from the outset for deployment in applied settings, targeting populations who are actively seeking to reduce or quit, and embedding the intervention within existing public health and clinical structures.

Cannabis and the CHanGE programme

Cannabis legalisation in Switzerland creates new public health challenges around cue-driven consumption and poses urgent questions about scalable, low-threshold interventions for regular users who wish to reduce their use. We are collaborating with the Geneva CHanGE programme (Cannabis and Health: a New Geneva Experiment), which provides a unique applied research framework for studying cannabis use patterns and interventions in the context of regulated access. Within this collaboration, we are investigating whether approach-avoidance and inhibitory control training can reduce cue-induced craving and consumption in regular cannabis users, and whether the individual difference moderators identified in our food research (notably Pavlovian learning style) predict treatment response in this population.

Alcohol and tobacco

In parallel, we are adapting the training platform for hazardous alcohol use and tobacco cessation. For alcohol, existing meta-analytic evidence supports the potential of inhibitory control training, but effect sizes are variable and the optimal parameters remain undefined. Our programme brings the mechanistic insights from the food domain (timing, semantic framing, endophenotype-based stratification) to bear on this question, with the aim of moving beyond generic protocols toward personalised, parameter-optimised interventions. For tobacco, where cue-reactivity is a well-established driver of relapse following quit attempts, we are exploring whether gamified training can serve as a low-cost adjunct to existing cessation support.

Integration with the MSI

Across all substance domains, this intervention programme is tightly coupled with our work on the Motivational Salience Index (MSI). The MSI provides the objective, implicit craving-monitoring layer that enables adaptive intervention delivery: by continuously tracking which substance cues retain motivational salience and which lose it over the course of training, the system can adjust cue selection and training intensity in real time. This closed-loop architecture is what distinguishes our approach from static training protocols and positions it for integration into clinical and public health workflows where ongoing monitoring is essential.

Translational model

All intervention studies are conducted at the University of Fribourg under Registered Report protocols, ensuring methodological rigour and independence from commercial interests. The gamified training platform is maintained by BeweLab SA (bewe.com), a University of Fribourg spin-off, which provides the technical infrastructure for large-scale, app-based data collection and, following peer-reviewed validation, supports the transfer of evidence-based intervention parameters into deployable digital health products. Clinical collaborations with partners in Geneva (CHanGE), Lausanne (CHUV), and the broader Swiss health network ensure that the research addresses real-world needs and that validated tools can be integrated into existing care pathways.

Selected references from the lab

Najberg, H. et al. (2023). Effects of gamified Go/NoGo training on sugar-sweetened beverage consumption. Registered Report.

Najberg, H. et al. (2021). A gamified inhibition training platform for food cue devaluation. Registered Report.

Tapparel, M. et al. (in press). Sign-tracking bias moderates Go/NoGo training-induced food devaluation. Registered Report.