The #psychology of #feedback: inspiration for ass-kicking #wearables
Posted on March 31st, 2015
03/30/2015 @TurnToTech, 184 5th Ave, NY
Brian Cugelman @Alertspark.com and @imcba.com spoke about ways to optimize feedback to maximize #behavior change. He sees a world eventually populated by inexpensive, ubiquitous #sensors that will gather data than can be used to mould behavior. One of the main challenges is to integrate
- Concentrate on sensor-base data capture.
- Data science –
- Interventions –
Currently, the most visible applications are health and fitness apps which monitor our behavior and give us feedback. The most successful apps apply evidence-based behavioral change programs which incorporate best health recommendations, monitoring behavior and give feedback on performance. Programs intend to educate and motivate using persuasive arguments to create healthy habits.
However, user falloff is steep and few follow the regimen for 3 months to create the target habits.
Brian described several areas of research that are relevant to better understand behavior to improve habit formation. These areas include
- Evidence-based design which shows that influence is maximized when there are neither too few arguments nor too many arguments presented to the audience
- Control theory argues that the loop of performance – monitoring – feedback can improve performance, but faces frictions along the way that can break the loop
- Persuasive design argues for two approaches to get a commitment
- Foot in the door – can you sign this petition, can you donate $100 – start small and go up
- Door in the face – would you donate $100, okay how about $10 – make a large request and then jump down – also used in ecommerce if someone abandons a shopping card (send a follow up email offering a discount)
- Tech that does not remind us to continue will fail. The more social interaction the better, even if the social interaction is just with a reminder program.
- The message is more persuasive the more we know about the person: this allows us to customize the message to make it more relevant to the user and better satisfy their needs and goals. Levels of interaction include
- Generic – we know little – can only give generic information (no feedback)
- Targeted – if know someone is part of a group – traditional segment analysis in advertising
- Personalized – superficial info (e.g. picture, name) – however, we should keep the message simple or it can become too complicated to understand. (and some combinations can hazardous to the user’s health)
- Tailored – know needs and desires.
Amazon uses a mix of these messages on its web page
- Targeting – message for android apps since they know he owns an Android phone
- Personalize – know his name
- Tailoring – know about specific interests
Brian then talked about novel ways to think of reinforcement and punishment.
He mentioned how Stickk motivates you by publically committing your to a goal and allows you and your friends to place money on the outcome.
He talked about how the punishment when a goal is not met should be carefully considered and is best designed to avoid undermining motivation, destroying confidence or pushing the user away from the technology that is designed to reinforce.
Brian also mentioned some references to better understand persuasion and feedback:
- Chaldini: Influence : the psychology of persuasion
- Stages of behavioral change
- BJ Fogg Persuasion design
- Clifford Nass Human-computer interactions
- Journal of Medical Internet Research