Comprehensive Study Guide on Aversive Control, Behavioral Interrelationships, and Stimulus Control

Aversive Control of Behavior and Nondiscriminated Avoidance

Nondiscriminated Avoidance Definitions: - In scenarios where there is no explicit warning stimulus preceding an aversive event, the contingency is termed nondiscriminated avoidance. - Sidman Avoidance (Free-Operant Avoidance): Named after Murray Sidman (1953), who first investigated this procedure. It involves a schedule where periodic shocks occur unless an operant response is emitted. - Calculated Intervals: - S–S (Shock–Shock) Interval: The time elapsed between consecutive shocks when no response occurs. - R–S (Response–Shock) Interval: The period of time by which a response delays the next scheduled shock. - Conditioning Effectiveness: Avoidance learning is optimized when the R–S interval is significantly longer than the S–S interval ( $R-S > S-S$ ).
The Avoidance Paradox: - Sidman observed that avoidance behavior is inherently cyclical: the more effective the response, the fewer shocks received; however, the fewer shocks received, the weaker the avoidance behavior becomes. - Maintenance of high-strength avoidance requires occasional negative reinforcement (shocks) to prevent extinction. - Comparison to Positive Reinforcement: Sidman stated that in positive reinforcement, "success breeds more of the same," whereas in avoidance, "success breeds failure" because the behavior stops unless another shock reinstates it.
Real-World Applications: - Public Health and Vaccination: Diseases function as negative reinforcers. Outbreaks (unpredictable/nondiscriminated) maintain vaccination. When diseases are eliminated, avoidance behavior (vaccination) declines, making new outbreaks probable. - Socio-Legal Compliance: Citizenship based solely on avoidance of jail results in cyclical temptations to break laws (speeding, tax cheating) unless occasional "shocks" (audits, check-stops) occur.

Determinants and Analysis of Avoidance Behavior

Shock Frequency and Molar vs. Molecular Accounts: - Molecular Account: Focuses on moment-to-moment variables, specifically the temporal intervals between shocks ( $S-S$ ) and responses to shocks ( $R-S$ ). - Molar Account: Suggests the overall reduction in shock frequency is the essential variable maintaining operant avoidance. - Herrnstein and Hineline (1966): 17 out of 18 rats showed avoidance responding even when shocks were not completely eliminated but their overall frequency was reduced.
Avoidance as "Impending Doom": - Hackenberg and Hineline (1987): Used a conditioned-suppression paradigm on rats. They found that responding for food reinforcement ( $FI\,3\,min$ ) was disrupted when periods of unsignaled shock avoidance ( $S-S = 5\,s, R-S = 20\,s$ ) either preceded or followed food sessions. - Suppression is often more severe when long-term aversive consequences are impending compared to immediate events.
Timeout (TO) from Avoidance: - Investigated by Perone and Galizio (1987), TO from avoidance functions as negative reinforcement. - Primary Driver: Research by Courtney & Perone (1992) indicates that reduction in response effort (suspension of behavioral requirements) is the key factor maintaining timeout behavior, rather than just the reduction in shock frequency.

Side Effects of Aversive Procedures

Behavioral Persistence: The successful use of punishment reinforces the punisher's behavior (via negative reinforcement), leading to more frequent use and increased side effects like counter-aggression.
Learned Helplessness: - Seligman and Maier (1967): Dogs exposed to inescapable, severe electric shocks failed to learn escape/avoidance later in a shuttle-box, effectively "giving up." - Human Analog (Hiroto & Seligman, 1975): Students exposed to inescapable loud noises failed to solve simple anagram problems. - Neuroscience of Helplessness: - Medial Prefrontal Cortex: Rich in $5-hydroxytryptamine\,(5-HT)$ (serotonin) receptors; modulates stress and "reward-negative" reactions. - Lateral Habenula (LHb): Implicated in "reward-negative" events; projects to the Ventral Tegmental Area (VTA). Increased excitatory synaptic response in LHb neurons correlates with depressive behavior.
Aggression Types: - Reflexive (Pain-Elicited) Aggression: Attack follows the presentation of aversive events (e.g., two rats fighting when shocked, even though neither is responsible). - Schedule-Induced Aggression: Extinction following a history of positive reinforcement induces aggression (e.g., hitting a vending machine that has taken money but failed to dispense). - Operant Aggression: Aggressive behavior shaped and maintained by negative reinforcement (removing the source of punishment via retaliation).
Social Disruption: Individuals attempt to escape or avoid the person or setting associated with punishment, leading to tardiness, truancy, and dropping out (Sidman, 2001).

Operant-Respondent Interrelationships

Dual Functions of Stimuli: A stimulus can simultaneously be a discriminative stimulus ( $S^D$ ) for an operant and a conditioned stimulus ( $CS$ ) for a respondent (e.g., a "Sit" command eliciting salivation while setting the occasion for sitting).
Instinctive Drift: Documented by the Brelands (1961), this refers to species-characteristic behavior patterns that interfere with operant conditioning (e.g., a raccoon rubbing coins together instead of depositing them because the coin has become a $CS$ for food-related rubbing patterns).
Sign Tracking: Approaching a stimulus that signals a biologically relevant event (food) rather than the food itself. Jenkins et al. (1978) showed dogs barking and prancing at a light/tone signal for food.
Autoshaping: - Brown and Jenkins (1968): Automatic shaping where a key light repeatedly precedes food, eliciting a peck (respondent) which is then reinforced (operant). - Nature of the behavior: Initially respondent, but becomes operant once the pecking is followed by food.
Visceral Conditioning (The Miller Experiments): Miller & DiCara (1967) used curare to immobilize rats and showed that heart rate could be specifically increased or decreased using Electrical Brain Stimulation (EBS) as reinforcement.

Taste Aversion and Preparedness

The Garcia Effect (Garcia & Koelling, 1966): Rats prepared to associate taste with illness, but not taste with shock. Birds (quail) were prepared to associate visual cues (blue water) with illness.
Key Features: - One-trial conditioning. - Sensitivity to novel tastes (neophobia). - Long CS–US delays (up to $12\,hours$ ).
Activity Anorexia: A cycle where food depletion increases physical activity (wheel running), and physical activity decreases the reinforcing value of food.
Bivalent Properties of Running: Wheel running before a flavor produces avoidance (CTA), while running after a flavor produces preference (CTP).

Adjunctive and Schedule-Induced Behavior

Interim Behavior: Actions like polydipsia (excessive drinking) that occur immediately after reinforcement on interval or time-based schedules.
Facultative Behavior: Behavior independent of the reinforcement schedule (e.g., grooming).
Terminal Behavior: Food-related activities occurring as the time for reinforcement nears.
Bitonic Function: Adjunctive behavior increases as the time between reinforcers grows from $2\,s$ to $180\,s$ , then drops off as the interval extends toward $300\,s$ .

Stimulus Control and Discrimination

Discrimination Index ( $I_D$ ): A measure of control exerted by $S^D$ and $S^{\Delta}$ . - $I_D = \frac{S^D\text{ rate}}{S^D\text{ rate} + S^{\Delta}\text{ rate}}$ - An $I_D$ of $0.50$ indicates no discrimination; $1.00$ indicates perfect discrimination.
Behavioral Contrast: - Positive Contrast: Rate of response increases in an unchanged component when reinforcement is decreased/extinguished in the other component. - Negative Contrast: Rate of response decreases in an unchanged component when reinforcement increases in the other. - Anticipatory Contrast: A strong contrast effect in an unchanged component that precedes a changed component ( $A$ in the $A \rightarrow B \rightarrow C$ sequence if $B$ is changed).
Generalization: - Generalization Gradient: Plotting probability of response against stimulus value (Guttman & Kalish, 1956). - Peak Shift (Hanson, 1959): The peak of a generalization gradient shifts away from $S^{\Delta}$ toward the other side of $S^D$ .
Errorless Discrimination & Fading: - Terrace (1963): Introducing $S^{\Delta}$ early and progressively (e.g., starting with a dark key and fading in green) results in learning without the emotional side effects of extinction.
Higher-Order Stimulus Control: - Matching to Sample (MTS): Includes Identity (IMTS), Relational (RMTS), and Delayed (DMTS). - Remembering as Discrimination: Geoffrey White (2002) argues remembering is a discrimination of events from a temporal distance. Accuracy is highest at the specific delay used during training (e.g., accuracy at $4\,s$ delay can be higher than at $0\,s$ if trained at $4\,s$ ).

Choice and Preference

Concurrent Schedules: Simultaneously available schedules (typically $VI\,VI$ ) used to study choice.
Changeover Delay (COD): A contingency (e.g., $3\,s$ ) that prevents frequent switching/alternation by delaying reinforcement after a switch.
The Matching Law (Herrnstein, 1961): The proportional rate of response matches the proportional rate of reinforcement. - $\frac{B_a}{B_a + B_b} = \frac{R_a}{R_a + R_b}$ - $B$ = Rate of response; $R$ = Rate of reinforcement.