Evaluation of Equivalence Relations: Models of Assessment and Best Practice

Due to changing age demographics in the United States, by 2050, an estimated 62.1 Americans will be over the age of 65 and the number of Americans with cognitive impairment, such as Alzheimer’s disease, will increase drastically as well (Alzheimer’s Association, 2014; Ortman, Velkoff, & Hogan, 2014). Once a diagnosis or behavioral indicators of cognitive impairment are present, it would be beneficial to apply a treatment package that promotes the maintenance or re-establishment of stimulus control in the environment. From a behavioral perspective, stimulus control aids in learning and memory through both respondent and operant conditioning. In the current study, stimulus equivalence training was completed and compared to cognitive and functional assessments scores with older adult participants with and without cognitive impairment as a systematic replication of Gallagher and Keenan (2009). Formation of equivalence relations after exposure to linear series (LS) training with 2 3-member stimulus classes across arbitrary, familiar, and stimuli from Gallagher and Keenan (2009) was compared via trials to criterion, accuracy per relation, and session length to scores on the Mini-Mental Status Exam (MMSE), Saint Louis University State Exam (SLUMS), and Barthel ADL Index. Several methodological changes were applied to a second study to examine the impact of increased programmed stimuli, training changes, and the use of one-to-many (OTM) and many-to-one (MTO) training structures with 3 3-member stimulus classes on equivalence formation. Composite performance scores were created for accuracy during LS, OTM, and MTO training. Non-parametric analyses were conducted between assessment and composite scores. The SLUMS and Barthel ADL Index were not correlated with any composite scores. However, MMSE scores and LS composite scores were correlated. The SLUMS was more sensitive to the detection of cognitive impairment as judged by classification and diagnoses. OTM and MTO composite scores also had a strong, positive correlation. Overall, more participants demonstrated higher levels of accurate responding during LS training than during OTM and MTO training. In contrast to previous research, only 25% of the sample demonstrated equivalence formation. Limitations and future research directions are discussed