The use of computational modeling for mapping the mind Marieke K. van Vugt 1 , [email protected] 1 Dept of Artificial Intelligence, University of Groningen, The Netherlands Modeling to disentangle effect of meditation on cognition Modeling to predict new effects of meditation on cognition • Detailed description of the cognitive process under study (Mehlhorn et al., 2012) • Verbal descriptions often ambiguous Why use modeling? • Decomposing a cognitive task into crucial cog- nitive operations • Defining it in equations or algorithms • Simulating the task on a computer • Matching parameters of the model to observed data • Changes in parameters indicate specific cogni- tive mechanisms What is cognitive modeling? Parameters: drift rate quality of informa- tion (inverse of men- tal noise) decision threshold response conserva- tiveness Ter non-decision time starting point bias The drift diffusion model of decision making van Vugt & Jha (2011) • 29 retreatants at Shambhala Mountain Center (ages 21–70) • One month - 6–10 hrs per day • Week 1 & 2: focus on breath • Week 3 & 4: widen focus and compassion • 29 age- and education-matched controls without meditation training tested one month apart mean RT var RT Why these changes? → Modeling! DDM shows reduction in perceptual noise Interaction between time and group: p =0.04 (non-parametric ANOVA) DDM shows reduction in perceptual noise data: Lutz et al. (2009) variability in drift rate → fluctuations of attention Decreased drift variability in dichotic listening cong inc neut cong inc neut 0 0.2 0.4 0.6 0.8 1 v Med Contr T1 T2 data: van den Hurk et al. (2010; submitted) Increased drift in attention network task Meditation decreases mental noise (More specific conclusions) Conclusions Can we simulate this on a computer? A conceptual model of meditation • Forces you to be precise • Connection to Western theories of cognition • Make predictions for transfer to cognitive tasks Why make a model of a meditating computer? • ACT-R is a cognitive architecture • Models cognition as a computer algorithm • Consists of modules reflecting cognitive re- sources: – visual/aural: perception – goal (ACC): keeping a goal in mind – declarative (frontal): declarative memory store – imaginal (parietal): working memory focus – motor/speech: produce responses – procedural (basal ganglia): proceduralizing sequences Introducing ACT-R cognitive architecture • Start with meditation instruction → put focus on goal “meditating” • Competition with a distracting “thought pump” process • How could it regain focus? Ideas? Outcome measures Fraction of time spent on the breath Length of distracted episodes Strength of productions (reflecting e.g., habits) Contents of distraction (pos vs neg memories) Outline of the meditating model production visual aural-location goal imaginal-action temporal imaginal vocal retrieval aural manual visual-location production visual aural-location goal imaginal-action temporal imaginal vocal retrieval aural manual visual-location production visual aural-location goal imaginal-action temporal imaginal vocal retrieval aural manual visual-location production visual aural-location goal imaginal-action temporal imaginal vocal retrieval aural manual visual-location production visual aural-location goal imaginal-action temporal imaginal vocal retrieval aural manual visual-location production visual aural-location goal imaginal-action temporal imaginal vocal retrieval aural manual visual-location production visual aural-location goal imaginal-action temporal imaginal vocal retrieval aural manual visual-location production visual aural-location goal imaginal-action temporal imaginal vocal retrieval aural manual visual-location production visual aural-location goal imaginal-action temporal imaginal vocal retrieval aural manual visual-location 0.600 0.650 0.700 0.750 0.800 0.850 0.900 0.950 1.000 1.050 1.100 1.150 1.200 1.250 1.300 1.350 1.400 1.450 1.500 1.550 1.600 1.650 1.700 1.750 1.800 1.850 1.900 1.950 2.000 2.050 2.100 2.150 2.200 2.250 2.300 2.350 2.400 2.45 2 production rec all -ne xt sta rt- tho ugh t-p ump rec all -ne xt sta rt- tho ugh t-p ump rec all -ne xt sta rt- tho ugh t-p ump visual aural-location goal imaginal-action temporal imaginal item item0 item item13 item item14 vocal retrieval item item5 item item3 item item5 ite m ite m5 aural manual visual-location Modeling the thought pump • Development of a computational model of meditation • Aim: comparing meditation model to task models • First: verify predictions for transfer to at- tentional blink • Next: make predictions for untested tasks (using Acttransfer - Taatgen, in press) • Ultimately: better understand why medi- tation helps people 0.6 0.7 0.8 0.9 2 4 8 lag T2|T1 (%) meditation FA OM experience low exp high exp van Vugt & Slagter (in preparation) Conclusions