Behavioural evaluation of methods for assessing fear responses in weaned pigs

Models of anxiety and fear of novelty were evaluated using correlations and principal component analysis. A total of 84 pigs (Landrace x Yorkshire) fr om nine different litters were subjected to a tonic immobility (TI) test at the age of 2.5 weeks, an elevated plus-maze (EPM) at the age of 6 weeks, a light/dark (L/D) exploration test at the age of 7 weeks and an open-field (OF) test at the age of 8 weeks. The first component from the principal component analysis had the highest c orrelation with number of entries into open arms in the EPM but was also hi ghly correlated to variables from the other three tests confirming a common aversion-related element in the four experimental tests. The second compon ent was negatively correlated with percent entries into and time spent on o pen arms in the EPM, but positively correlated with the number of entries i nto closed arms in the same test, number of lines crossed in the OF and tim e spent in the lit compartment of the L/D test. The last point illustrates a negative relationship between "anxiety" and "activity" in the EPM and OF. To achieve purer measures of fear of novelty and activity in the tests, th e components were rotated using the Varimax criterion. The rotated factor p attern demonstrated a simple structure where variables related to ''anxiety '' or ''fear of novelty'' (i.e., percent entries into open arms and time sp ent on open arms of the EPM) had the highest loading on factor 1, whereas v ariables related to activity (i.e., number of entries into the closed arms in the EPM, number of lines crossed in the OF and time spent in the lit com partment of the L/D test) had the highest loading on factor 2. TI duration loaded more strongly on factor 1 ("fear of novelty") than on factor 2 ("act ivity"), but did not represent any pure measure of either fear of novelty o r activity. In conclusion, all of the test variables were related to one another. Open- arm avoidance represented the purest measure of fear of novelty, whereas en tries into closed arms and number of lines crossed in the OF were the pures t measures of activity. The EPM appeared to provide the best way to separat e the fear of novelty and activity-related elements, indicating that the EP M may be a useful behavioural model of fear of novelty or avoidance in pigs . (C) 2000 Elsevier Science B.V. All rights reserved.