Application of three different coaching strategies through a virtual coach for people with emotional eating: a vignette study

Abstract

Background Around 13% of the world’s population suffers from obesity. More than 40% of people with obesity display emotional eating behaviour (eating in response to negative emotions or distress). It is an alternate to more effective coping strategies for negative emotions. Our study explored the opportunities for helping adults with emotional overeating using a virtual coach, aiming to identify preferences for tailored coaching strategies applicable in a personal virtual coach environment. Three different coaching strategies were tested: a validating, a focus-on-change, and a dialectical one – the latter being a synthesis of the first two strategies. Methods A qualitative study used vignettes reflecting the two most relevant situations for people with emotional eating: 1. experiencing negative emotions, with ensuing food cravings; and 2. after losing control to emotional eating, with ensuing feelings of low self-esteem. Applied design: 2 situations × 3 coaching strategies. Participants: 71 adult women (Mage 44.4/years, range 19–70, SD = 12.86) with high scores on the DEBQ-emotional eating scale (Memo 3.65, range 1.69–4.92, SD = .69) with mean BMI 30.1 (range 18–46, SD = 6.53). They were recruited via dieticians’ practices, were randomly assigned to the conditions and asked how they would face and react to the presented coaching strategies. Data were transcribed and a thematic analysis was conducted. Results Qualitative results showed that participants valued both the validating coaching strategy and the focus-on-change strategy, but indicated that a combination of validation and focus-on-change provides both mental support and practical advice. Data showed that participants differed in their level of awareness of the role that emotions play in their overeating and the need for emotion-regulation skills. Conclusion The design of the virtual coach should be based on dialectical coaching strategies as preferred by participants with emotional eating behaviour. It should be tailored to the different stages of awareness of their emotions and individual emotion-regulation skills.

The Use of Artificial Intelligence–Based Conversational Agents (Chatbots) for Weight Loss: Scoping Review and Practical Recommendations

“…In total, 12 studies mentioned the use of a behavior change framework to guide AI chatbot development ( Multimedia Appendix 4 ). A total of 3 studies used motivational interviewing [ 26 , 43 , 45 ]; 2 studies used cognitive behavioral therapy [ 37 , 45 ]; and others used mindfulness-based stress reduction [ 26 ], dialectic behavior therapy [ 41 ], efficiency model of support [ 39 ], and control theory by Carver and Scheier [ 34 ]. Moreover, 5 studies [ 29 , 33 ] referenced the use of taxonomy of behavior change techniques, whereas the remaining (9/23, 39%) studies did not specify the use of a structured behavior change framework (ie, briefly mentioned the incorporation of behavior change techniques such as goal setting, problem solving, and self-monitoring).…”

“…Core functions of the chatbot were to provide personalized weight loss recommendations (20/23, 87%) [ 25 , 26 , 29 - 34 , 37 - 39 , 41 - 46 ] and motivational messages (18/23, 78%) [ 25 , 26 , 29 - 33 , 37 - 39 , 41 - 43 , 45 , 46 ] ( Multimedia Appendix 5 [ 25 , 26 , 29 - 37 , 39 - 47 ]). Only 26% (6/23) studies [ 30 - 32 , 36 , 40 , 43 ] mentioned the use of gamification to enhance user engagement, and 26% (6/23) studies [ 37 , 39 , 40 , 42 , 43 , 45 ] mentioned the use of sentiment analysis to provide emotional support through emotionally appropriate messages ( Multimedia Appendix 5 ).…”

“…Some techniques used were the deployment of humanoid robots [ 43 ] and embodied chatbots [ 26 ] that were capable of displaying visual social cues such as eye contact and hand gestures. Other techniques used in text-based chatbots were the delivery of colloquial, personally and culturally appropriate conversational tones and content [ 26 , 30 ]; emojis to emulate human emotional expressions [ 30 - 32 ]; positively framed words [ 32 ]; citations of credible information sources [ 26 , 33 ]; and validation (eg, acknowledgments and compliments) of not only behaviors but also thoughts and feelings [ 41 ]. Participants of the included studies were also found to have appreciated the personification of the chatbot (eg, funny, animated, empathetic, or playful) [ 25 , 29 , 31 , 32 , 37 ] and the provision of real-time, fast, and reliable recommendations [ 26 , 29 , 38 ].…”

“…In contrast, users mentioned concerns regarding privacy and accountability [ 43 , 46 ], the inconvenience of having the chatbot on a limited number of third-party platforms (eg, only Telegram that one may not use) information [ 31 ], message or question overload that causes user fatigue [ 25 , 31 ], transparency about the app objectives and information sources [ 31 ], and appearing too robotic (eg, speaking too slowly in a robotic voice). Users also suggested that the chatbots should probe further to explore emotions and action plans instead of prescribing them [ 41 ]. Most strikingly, users suggested the incorporation of progress-based recommendations, rewards for goals achieved (ie, gamification), and integration with other tracking devices and appliances through IoT [ 25 , 26 ].…”

Preferences for coaching strategies in a personalized virtual coach for emotional eaters: an explorative study

“…To best meet the emotional eaters’ need for right-on-time coaching ( Gravenhorst et al, 2015 ; Dekker et al, 2020 ), we have choosen the two tipping points, the two proto-typical problem situations ( Safer et al, 2009 ; Dol et al, 2021 ), at which interventions might be most effective:…”