What is the CBS Superlab?
With the release of the ACBS Task Force Report on the Strategies and Tactics of Contextual Behavioral Science Research, high-level discussion around how to roll out the proposed recommendations is necessary. The CBS Superlab is an international research lab meeting held once a quarter via Zoom. CBS SuperLab Meetings are typically held at 3pm EST. These hour-long quarterly meetings will involve:
• A research presentation delivered by a CBS lab that showcases ongoing advances, developments, and innovations in the field of CBS. Each presenter will be invited to share resources relating to their presentation (e.g., PowerPoint slides, handouts, software packages) that will be made available to all attendees.
• A group discussion among all attendees that focuses on both the presentation and means of addressing the Task Force’s recommendations.
All ACBS members are invited to attend. All CBS research labs are invited to participate. To be considered a CBS lab, your lab details must be included on the www.contextualscience.org website. To attend, register here and join the Superlab listserv to continue the conversation.
Research labs may submit to present here.
Superlab with Robert Johansson, PhD, Department of Psychology, Stockholm University, Sweden - April 24, 2024 3pm EST (You will find the recording here).
Title: "Artificial General Intelligence from a contextual behavioral science perspective."
Abstract: Artificial General Intelligence (AGI) is a branch of AI research aiming to create "thinking machines" - computer systems that have a general-purpose ability, as distinguished from most of today's systems that are trained for special-purpose tasks. One example of an AGI system is Pei Wang's Non-Axiomatic Reasoning System (NARS). Due to its design, including an experience-grounded semantics and its focus on adaptation with insufficient knowledge and resources, there is a natural overlap with much of contextual behavioral philosophy and psychology.
Since 2018, our lab has been running a research program where we investigate the relational reasoning abilities of NARS, using methods from behavioral psychology and RFT. We have proved NARS to be able to learn with both classical and operant conditioning in a way that is very similar to human and non-human operant learning. In addition we have also more recently showed that NARS is able to learn generalized identity matching, and, after a sufficient history of multiple-exemplar training, perform in accordance with symmetry.
This talk will give a brief introduction to the AGI research field with a particular focus on NARS, Furthermore, we will describe the research we have been carrying out. Finally, we will argue why we believe this kind of research could be a valuable contribution to ACBS, and in line with several recommendations of the ACBS Task Force.
Superlab with Phillip Klein - July 17, 2024 3pm EST (Register for this meeting and all of the quarterly Superlab meetings here).
Title: "From Network Theory to Process-Based Therapy: a practice-oriented research presentation."
The network theory of mental disorders stipulates that mental disorders result from a dynamic interaction between symptoms of mental disorders and other variables. This implies that symptoms of mental disorders are not passive indicators of an underlying causal mechanism and might have profound implications for treatment. These implications form the basis of process-based therapy which states that psychological treatments should no longer be defined by treatment protocols that target the putative causal mechanism of a certain mental disorder but rather be defined by therapeutic strategies aimed at changing the network dynamics from a disease state to a healthy state. It is unclear however, which networks characteristics have to be targeted to bring about this change. In this session, we will first introduce network theory and its connection to process-based therapy Jan Philipp Klein (Lübeck University, Germany) and then present three studies:
- Lea Schumacher (University Medical Center Eppendorf, Hamburg, Germany) will present person-specific networks estimated from data of a randomized-controlled psychotherapy trial that sheds light on the question which changes in these networks best predict outcome: changes in symptom associations or changes in centrality parameters?
- Jana Bommer (Trier University, Germany) will present intra-individual networks estimated from data of a large population-based study to examine whether vicious cycles that can be identified on a group level are present on the individual level and if these vicious cycles actually confer an unfavorable outcome.
- Finally, Nele Assmann (Lübeck University, Germany) will present cross-sectional networks of patients who are seeking treatment for depressive disorder, anxiety disorder, somatic symptom disorder or eating disorder to ascertain which core psychological processes are associated with these syndromes.
By the end of the session, attendees will have a clearer understanding of the state of the research on network theory's contribution to the practice of psychotherapy.