On the Similarity Between the Reinforcing and the Discriminative Properties of Intracranial Self-Stimulation

Abstract

Rats work very hard for intracranial self-stimulation (ICSS) and tradeoff effort or time allocation for intensity and frequency parameters producing a sigmoidal function of the subjective reward magnitude of ICSS. Previous studies using electrical intracranial stimuli (ICS) as a discriminative cue focused on estimating detection thresholds or on the discrimination between intensities. To our knowledge, there is no direct comparison of the reinforcer tradeoff functions with the discriminative functions. Rats were trained to press and hold the lever for ICSS using the maximum reinforcing intensity below motor alterations or avoidance behavior. First, rats were trained to hold the lever for 1 s; after stability, they undergo trials where intensity or frequency was decreased on 0.1 log step. Thereafter, they undergo further training with a hold of 2 and later of 4 s to determine tradeoff with intensity or frequency. The same rats were trained on a discrimination task where the previously used ICSS signaled a lever where a 1 s hold response was followed by a reinforcing ICSS; on randomly alternating trials, a −0.6 log ICS signaled an alternate lever where a similar hold response led to a reinforcer. After mastering discrimination, generalization tests were carried out with varying intensity or frequency. Rats completed training with 2 and later 4 s hold response. After the completion of each task, the rats had different doses of a pimozide challenge while their intensity and hold-down requirement were varied. With regards to the rats’ tradeoff response time allocation as a function of intensity or frequency, sigmoid functions were displaced to the right when long responses were required. Rats that learned the discrimination task attained a discrimination index of 90–98%. Discrimination accuracy decreased slightly with the increase of hold requirement, but generalization gradients were not displaced to the right as a function of the response requirement. Pimozide induced a dose-dependent displacement of the time-allocation gradients, but it did not affect the generalization gradients. It is concluded that rats integrate response requirements as part of the reinforcement tradeoff function, but the response cost is not integrated into the discriminative function of ICSS.

Comparison of progressive hold and progressive response schedules of reinforcement

“…In previous work that examined time allocation we observed deceasing time allocation as work requirement increased (e.g. Trujillo-Pisanty et al 2020; Velazquez-Martinez et al 2022). However, it should be noted that in previous studies available time for responding (usually enough time to collect a fixed number or reinforcers) was controlled by the schedule, but in the PR schedule, both PS R P duration and IS R T was controlled by the subject.…”

“…It has been suggested that a more basic unit of measurement than actual numbers of discrete responses may be time allocated to an activity (Baum and Rachlin 1969); then, the hold-down response that excludes any other behavioral responses may truly reflect WT. Hold-down responses had been used successfully by the Reward Mountain Model to have an exact estimation of time allocation and, as commented previously, demonstrated that time allocation is related to the joint function of reward strength (frequency, amplitude or train duration) and the cumulated time or effort to harvest rewards in various electrical-(Arvanitogiannis and Shizgal 2008; Hernandez et al 2010; Trujillo-Pisanty et al 2011; Hernandez et al 2012; Trujillo-Pisanty, Conover, and Shizgal 2013; Solomon et al 2015; Solomon, Conover, and Shizgal 2017; Trujillo-Pisanty et al 2020; Velazquez-Martinez et al 2022) and optical-ICSS (Pallikaras et al 2022) procedures. Worth to notice is that even the shortest hold requirements rarely were completed with a single lever-holding response, but rather lever was released for very brief periods several times within the IS R T (Rider and Kametani 1984).…”

“…requirements, but later, it increased as the work requirement increased. In previous work that examined time allocation we observed deceasing time allocation as work requirement increased (e.g Trujillo-Pisanty et al 2020;Velazquez-Martinez et al 2022…”

“…Gallistel et al 1991; Mark and Gallistel 1993; Simmons and Gallistel 1994). Diverse work with the RMM demonstrated that the amount of work (time allocation) a subject would spend working is related to the joint function of reward strength and opportunity cost (cumulated time or effort to harvest rewards) (Arvanitogiannis and Shizgal 2008; Hernandez et al 2010; Trujillo-Pisanty et al 2011; Hernandez et al 2012; Trujillo-Pisanty, Conover, and Shizgal 2013; Solomon et al 2015; Solomon, Conover, and Shizgal 2017; Trujillo-Pisanty et al 2020; Velazquez-Martinez et al 2022). Previously, there has been attempts to use continuous lever depression as measure of effort in progressive schedules (Progressive Hold-down, PH) where the step increase changes within a session (Bailey et al 2015; Gulotta and Byrne 2015) or between successive sessions (Rider and Kametani 1984, 1987; Peck and Byrne 2016).…”