|
| Computers, Decimals, and Classification, Oh My! |
| Saturday, May 26, 2018 |
| 12:00 PM–12:20 PM |
| Marriott Marquis, Grand Ballroom 10-13 |
| Area: PCH |
| Chair: Martti T. Tuomisto (Faculty of Social Sciences, University of Tampere) |
| |
CANCELED: Behaviour Analysis, Computational Modelling, and the Imitation Game |
| Domain: Theory |
| DON (YUHAN) LI (The University of Auckland) |
| |
| Abstract: Behaviour is a high-dimensional multivariate construct. One way to provide a complete representation of the behaviour that we measure is to assert identity; i.e., to take the event record as the best representation of itself. But because of the high dimensionality of the event record, we often compute dependent variables to summarise particular dimensions of the event record. Without knowing how to summarise the event record with a finite set of dependent variables without loss of information, a complete and unified theory of behaviour would require the ability to contend with any dependent variable that we may compute from the event records of interest. A way for us to understand the demands of a unified theory of behaviour is to use Turing's (1950) Imitation Game; whereby a theory of behaviour is unified and complete if its outputs are indistinguishable from those of a given subject under some experimental contingency regardless of how we choose to examine the difference between subject and model. Through the Imitation Game, we explore the extent to which three classes of theory (qualitative, finite systems of algebraic equations, and computational models of responding) could serve as, or provide a pathway to, a unified theory of behaviour. |
| |
Principles and Functions of Behavior in the Classification of Behavior Problems Using the "Decimal System" |
| Domain: Theory |
| MARTTI T. TUOMISTO (University of Tampere) |
| |
| Abstract: The "Decimal System" of Behavior and Life Problems (DSBLP) was developed in 2010-2013 for the classification of behavior problems and problems with life circumstances according to behavior-analytic concepts and principles to get a comprehensive and functional tool to guide goal setting and interventions in behavior analysis. DSBLP also includes a "mirror image" of the problem classes defined: respective behavior assets and strengths. DSBLP has nine main problem classes with subclasses and a zero class for non-problems. Seven classes are pure behavior problems of which only one class can be a primary problem at one time. Of the two problem classes, environmental and biological problems ("life problems"), one can be a primary problem simultaneously with one of the behavior problem classes. Consequently, other problem classes can be secondary problems. The DSBLP enables tailor-making individual treatments according to the goals derived from the system. The complaints of a client are first assessed using "chain analyses" of behavior that are formulated based on hierarchical decision trees included in an applied or clinical interview. The interview renders possible different types of functional assessments and analyses including experimental functional analysis. In this presentation, I will go through the different problem classes and their behavioral functions. |
| |
|
| |