The Science of Predicting and Changing Human Behavior - Episode 562 (M02)

·39 min·3 clips

Algorithms are becoming so powerful at assessing and controlling the human mind that some see it as a major threat to humanity.

The episode begins with an ad for Libsyn Ads and a host introduction, setting up the February 2025 installment of the Read to Lead podcast. Jeff Brown introduces the guest, Sondra Matz, author of 'Mind Masters', noting that his online community unanimously selected her book for discussion. Matz opens by framing the data debate through the analogy of growing up in a small German village, where community involvement offered both support and intrusion, similar to today's algorithms. She explains how digital footprints, such as GPS data and social media activity, provide insights into psychology, emphasizing the unintentional 'behavioral residue' we leave. The conversation shifts to personality, with Matz arguing that traits are not fixed but fluctuate with context, and algorithms can capture this dynamism to move beyond simplistic categorizations. She distinguishes between ethical persuasion and manipulation, defining the latter as actions not in a person's best interest or without their understanding. Matz highlights mental health as a promising application, where smartphone data can detect early signs of depression through changes in behavior and language, such as increased use of first-person pronouns. She critiques common privacy fallacies like 'I have nothing to hide' and discusses the lack of reciprocity in data relationships. To address these issues, Matz advocates for federated learning, which allows data processing on devices without central collection, reducing security risks. She also proposes data co-ops, inspired by village reciprocity, where communities pool data for mutual benefit, such as expecting moms sharing health information. The episode ends with Matz identifying federated learning as the most urgent step to eliminate trade-offs between privacy and convenience, followed by host outro and community promotion.

As heard by us

Personality is most revealing when behavior, context, and data are considered together.

Human behavior comes through here as fluid rather than fixed, and that gives the episode its clearest point of interest. The conversation frames personality as a range of states shaped by circumstance, not a single permanent label: a person who is usually highly extroverted…

Read the full review in PlayNext →

Why you'd press play

If you want to understand why your personality can change with the room you are in, press play.

Read the full recommendation in PlayNext →

Listen to the show on

Apple Podcasts Spotify

Play clips from this episode More from Read to Lead Podcast