libitum weight during the study; (2) the probability that a new ad libitum weight is desirable 
because of the age of the animal at the time of original determination (or because of seasonal 
variations in weight with adult male squirrel monkeys); (3) the extent to which particular 
species tend to "waste" or scatter food (e.g., monkeys) under free-feeding; and (4) whether 
there are problems created by abrupt shifts between restricted and unrestricted feeding (e.g., 
bloat in some monkeys). 
REGULATING ACCESS TO FLUID 
When water is used to maintain stable responding, access to water outside the experimental 
session needs to be controlled. The influence of varying amounts of water restriction on 
operant performance has been described (Hughes et al., 1994). In addition, some other liquid 
reinforcers (e.g., fruit juice with monkeys) under certain conditions (e.g., procedures that 
require long sessions with many reinforcer deliveries) may also maintain performance most 
reliably when access to water is controlled. 
Fluids have advantages over solid food reinforcers for behavioral procedures that might 
require that the animal's head be kept in a particular position (e.g., psychophysical studies or 
studies that monitor brain activity in awake, behaving organisms). In such cases, the fluid 
may be delivered through a solenoid-operated sipper tube positioned at the animal’s mouth. 
A particular advantage of fluids when an experiment involves neuronal recordings with 
microelectrodes is that chewing or crunching movements of the teeth or jaws does not occur 
when the animal is consuming the reward. 
Animals physiologically tolerate a lack of food better than a lack of water. Determining 
parameters of water restriction that do not produce dehydration or excessive weight loss 
requires careful consideration. Animals need not be at risk if intervals of fluid access and 
total amounts of fluid obtained are appropriate to the species (ILAR, 1995; Toth and Gardiner, 
2000 ). 
Some studies using fluid delivery to maintain a behavioral performance require that the 
animal earn its daily fluid requirement during the experimental session, and these sessions 
typically are multiple hours in length. Other studies use shorter sessions, but provide a period 
of supplemental access to water shortly after the session. On days when sessions are not 
conducted, animals should receive a period of access to water, unless there is strong 
experimental justification for not doing so (e.g., when duration of fluid restriction is an 
independent variable). 
The main disadvantage of fluid control in very small animals is the risk of rapid dehydration if 
the animal fails to receive its daily water requirement. A good system of daily monitoring 
procedures is essential under such protocols. Records should be kept of the amount of fluid 
47 
