|
| Int'l Symposium - Transformation of Function: Empirical Investigations |
| Monday, May 31, 2004 |
| 3:00 PM–4:20 PM |
| Fairfax A |
| Area: EAB/VBC; Domain: Applied Research |
| Chair: Ian T. Stewart (National University of Ireland, Galway) |
| Abstract: . |
| |
| A Derived Transformation of "Expectancy" Functions: Developing a Relational Frame Model of Irrational Fears |
| SINEAD SMYTH (National University of Ireland, Maynooth), Dermot Barnes-Holmes (National University of Ireland, Maynooth) |
| Abstract: Numerous studies have reported the transformation of functions in accordance with derived stimulus relations. Some of these studies have attempted to develop models of human psychopathology by investigating the derived transfer of elicited fear responses. In these studies, one member of an equivalence class is directly paired with an aversive UCS (e.g., electric shock), and then remaining members of the class are presented in the absence of the UCS. Typically, the other equivalence class members produce levels of arousal similar to that produced by the directly paired stimulus. One interpretation of this outcome is that all the equivalence class members come to control a derived “expectancy” response (i.e., the participant expects to receive a shock whenever an equivalence class member is presented). Although this expectancy-based interpretation of the derived transfer of elicited fear functions seems reasonable, it has not been systematically tested under laboratory conditions. The current paper presents a research program that set out to develop experimental procedures for the investigation of the derived transfer of expectancy functions. The research highlights the role of multiple-exemplar training in establishing and maintaining derived expectancies and appears to provide a more valid model of irrational fears than previous research in this area. |
| |
| A Transformation of Stimulus Functions Based on an Analogical Relational Network |
| FRANCISCA LOPEZ RIOS (University of Almeria, Spain), Serafin Gomez-Martin (University of Almeria, Spain), Dermot Barnes-Holmes (National University of Ireland, Maynooth) |
| Abstract: The paper reports a study that attempted to show that a derived relation might acquire a new function from another similar derived relation, thereby providing a model of how analogy can be the basis for the transformation of stimulus functions. Specifically, subjects were exposed to five experimental phases: 1) training to form four equivalence classes (i.e., A1-B1-C1, A2-B2-C2, A3-B3-C3, A4-B4-C4); 2) tests for equivalence within the classes; 3) training Function 1 (F1) to relations composed of same class stimulus (e.g., B1A1, B2A2, etc.) and training Function 2 (F2) to relations composed of different class stimulus (e.g., B1A2, B2A1, etc.); 4) tests for transformation of stimulus functions to new relations (e.g., would C1A1, C2A2 and the like acquire F1 and C1A2, C2A1 and the like acquire F2?); 5) test to demonstrate analogy within the context of matching to sample (e.g., with C1A1 as sample would subjects choose C2A2 when C2A2 and C3A4 are the comparisons?). These results may shed light on how psychological functions are transformed in accordance with analogical relations. |
| |
| Equivalence-Equivalence and Transfer of Function: Further Investigations |
| DONNLA HUGHES (National University of Ireland, Galway), Ian T. Stewart (National University of Ireland, Galway), Dermot Barnes-Holmes (National University of Ireland, Maynooth), Louise A. Mchugh (National University of Ireland, Maynooth) |
| Abstract: Previous studies modeled analogy as equivalence-equivalence responding and demonstrated transfer of function in accordance with equivalence-equivalence. This study investigated the effect of variety of exemplars of transfer training on transfer of function through equivalence-equivalence. Adult subjects were first divided into three groups, all of which were trained and tested for the formation of four 3-member equivalence relations: A1-B1-C1, A2-B2-C2, A3-B3-C3, and A4-B4-C4. They were then tested for equivalence-equivalence (e.g., matching B1/C1 to B2/C2 rather than B3/C4). For Group 1, function F1 was then trained to 1 compound stimulus composed of two directly trained same-relation stimuli, and function F2 to 1 compound composed of directly trained different-relation stimuli. Subsequent tests probed for F1 in the presence of compounds containing two equivalent stimuli and F2 in the presence of compounds containing two non-equivalent stimuli. Group 2 received the same number of transfer training trials, but for this group, F1 was trained to 2 different exemplars of trained same-relation compounds and 2 different exemplars of trained different-relation compounds. For Group 3, F1 was trained to 4 exemplars of same-relation compounds and 4 exemplars of different-relation compounds. Results showed that transformation of function was more likely given greater variety of transfer training exemplars. |
| |
| Relations between Multiple-stimulus Relations and Transformation of Functions using Respondant-type Training: a Model of Analogy |
| HILARIO MESA MANJON (University of Almeria, Spain), Serafin Gomez-Martin (University of Almeria, Spain), Dermot Barnes-Holmes (National University of Ireland, Maynooth) |
| Abstract: This paper reports an attempt to demonstrate analogical reasoning with relations other than those involved in a frame of coordination, and to test for the transformation of function between analogical relations. Different subjects were exposed to a complex training protocol in which contextual cues for direction (bottom-to-top or top-to-bottom), more than, less than, same and different were established with the aim of generating a series of stimulus networks (e.g., A>B>C; D>E>F; G>H>I). Once these networks were created and analogical reasoning was demonstrated, new stimulus networks (J>K>L ; LL>M>N ; O>P>K) were generated to demonstrate transformation of stimulus functions with new relations. Furthermore an more elaborated respondent-type training procedure was used to establish these new relations. |
|
| |