Pigs acquire, through learning and evolution, expectations of their environ
ment. Frustration of expectations results in motivation to change these con
ditions and is therefore adaptive. Initially, frustration should produce pr
oblem solving behaviour. If these responses are unsuccessful, other behavio
urs, reflecting general frustration should be elicited. Our purpose was to
study both types of responses to frustration in grower pigs. Eighteen growe
r pigs were fasted for 1, 2 or 3 h each morning by removing the feeders. On
training days, Monday and Tuesday of each week, full feeders were returned
to the pigs. On test days, Wednesday to Friday of each week, nonfunctional
feeders were placed in the pens. Pigs were videotaped for 2 h following re
turn of the feeders and behaviour on training and test days was compared. T
wo types of feeders were utilized, non-lidded (O) and lidded feeders (L). N
on-lidded feeders, when placed in the pens were empty while lidded feeders
had the lids bolted down to prevent access to feed. Frustration periods wer
e compared to control periods by split-plot-in-time analysis (p < 0.05). A
6 X 6 Latin Square design incorporated six treatments (two feeder types X t
hree fast periods) over 6 weeks. Both single pigs (S) and pairs (P) were ob
served. During the period of frustration both S- and P-pigs increased their
general activity level. This was reflected in a decrease in lateral recumb
ency(S: -12.7% and P: -9.9%) with an increase in sternal recumbency (+10.8%
) and sitting (+3.0%) in the S-pigs and an increase in sternal recumbency (
+5.2%), sitting (+3.5%) and playing (+0.7%) in the P-pigs. Oral manipulatio
n of the feeder by S-pigs was more pronounced during the first hour than th
e second hour in L-feeders (7.1% vs. 5.2%) but not in O-feeders (6.1% vs. 6
.5%) indicating an increase in activities related to problem solving. Frust
ration increased general oral activity including increased oral manipulatio
n of other pigs (P: 1.9% to 5.6%), and other objects (S: 8.8% to 28.4% and
P: 6.9% to 14.9%). Serum cortisol levels were determined prior to, and at t
he conclusion of the experiment, as an indication of stress during the 6 we
eks of on-and-off frustration. Although all pigs were exposed to all treatm
ents, those pigs exposed to L-feeders during the final week had higher cort
isol levels (92.3 nmol/l) than those exposed to O-feeders (56.9 nmol/l). Fr
ustration, as expressed in increased oral activity, did not substantially d
iminish over a 6-week period although general activity levels tended to ret
urn to normal. (C) 1999 Elsevier Science B.V. All rights reserved.