The Sometimes Opponent Processes Model (SOP): An In-Depth Examination

In the intricate tapestry of life, the relationship between organisms and their environments is a dynamic interplay that shapes behavior, learning, and emotional responses. This complexity underscores our understanding of how living beings adapt to ever-changing surroundings through various mechanisms of associative learning. Emotions serve as vital indicators of these interactions, where each stimulus encountered can provoke not just immediate reactions but also a series of aftereffects that reverberate over time. Understanding this sophisticated dance between stimuli and responses provides profound insights into both individual experiences and broader behavioral patterns.

At the heart of this exploration lies the Sometimes Opponent Processes (SOP) Model—a theoretical framework developed to elucidate how organisms navigate their environments by processing information in nuanced ways. By examining the dual activation states inherent to memory—primary (A1) and secondary (A2)—the SOP model reveals how initial strong emotional responses are often countered by slower, opposing processes that strive for equilibrium within an organism’s internal landscape. Through repeated interactions with stimuli, these processes evolve.

They impact not only our ability to learn. In addition, they also influence phenomena such as tolerance and withdrawal in response dynamics. As we delve deeper into the SOP model, we uncover layers of understanding about motivation and emotion intertwined with associative learning principles—an essential key to unlocking the mysteries behind behavior in an ever-adaptive world.

Key Definition:

The Sometimes Opponent Processes (SOP) Model, proposed by Richard L. Solomon and John D. Corbit in 1974, is a theory of motivation and emotion that explains why emotions and pleasure/displeasure responses often involve an initial strong reaction followed by an opposing, less intense, and prolonged after-reaction. It posits that every strong emotional or hedonic experience (the a-process) initiates an opponent, counteracting process (the b-process).

Introduction: Exploring Associative Learning through Computational and Psychological Lenses

The Sometimes Opponent Processes Model (SOP) has been a significant theoretical advancement in the field of associative learning. Introduced by Allan R. Wagner in 1981, SOP offers a mathematical and conceptual framework for understanding how organisms learn associations between stimuli over time (Wagner, 1981). By addressing the nuances of memory activation states and their role in learning, SOP has provided a robust explanation for a wide range of experimental phenomena previously unaccounted for by simpler models. This article explores the core components of SOP, its empirical support, extensions, and its place in contemporary learning theory.

Homeostasis and the SOP

Furthermore, SOP draws inspiration from the broader Opponent-Process Theory of Motivation, which helps explain how organisms strive for internal balance, or homeostasis, when experiencing strong emotional or physical events. This theory suggests that an initial intense reaction (an ‘a’ process) automatically triggers a slower, opposing reaction (a ‘b’ process) that works to counteract it.

For instance, the immediate unpleasantness of a shock (the ‘a’ process) is eventually followed by a sense of relief (the ‘b’ process). With repeated exposure to the shock, this ‘b’ process strengthens, making the original shock less impactful over time. SOP applies this principle to memory, explaining habituation—the gradual decrease in response to a repeated, unimportant stimulus—as a “priming” effect. This means that when a stimulus is presented repeatedly, its representation in memory becomes “primed” or pre-activated, which in turn reduces the amount of new processing it requires.

This priming can result from recent exposures. This explains short-term habituation. In addition, it can also result from the environmental context becoming associated with the stimulus. This explains long-term habituation. Thus, SOP provides a detailed theoretical framework for how learning shapes an organism’s responses to environmental cues by managing the complex internal states that strive for balance.

Theoretical Context of SOP

Wagner’s Sometimes Opponent Processes (SOP) model is a significant theoretical framework in the field of learning and memory, particularly concerning habituation and associative learning. Its theoretical context is rooted in several preceding and contemporary ideas, which it integrates and quantitatively elaborates upon.

Foundation in Priming Theory

Priming theory is particularly explained by Allan Wagner’s Sometimes Opponent Processes (SOP) model. It shows how our previous experiences with a stimulus can influence our subsequent processing of and response to it. Cues associated with that stimulus also play a role in this influence. At its core, the theory posits that the more a stimulus’s representation is already “pre-represented” or “primed” in short-term memory (STM), the less effective that stimulus will be in eliciting its full primary response or in forming new strong associations.

This happens because elements of the stimulus’s representation are either in a “primary active” (A1) state or a “secondary active” (A2) refractory state, which limits their ability to be freshly activated to drive a response or new learning (Wagner, 1981, p. 6).

Wagner’s SOP model expands on this idea. It explains that when an event is already present in our short-term memory, additional similar information isn’t as impactful. In other words, if we’re already familiar with something, experiencing it again doesn’t grab our attention or provoke a reaction quite like it would if we were encountering it for the first time (Uribe-Bahamonde et al., 2019).

Short-Term Habituation

Priming is a common daily experience. Imagine a new, repetitive sound, like a leaky faucet drip or a new refrigerator hum. Initially, you might actively notice the sound, finding it distracting or even annoying. However, with repeated, un-consequential exposure over a short period, you gradually stop consciously perceiving it or reacting to it. This occurs because the stimulus’s representation enters a “secondary activity” (A2) state, which prevents it from fully re-engaging your primary attention (A1 state) upon subsequent presentations, leading to a decline in your response.

Long-Term Habituation

Priming explains why you might be less bothered by certain stimuli even after a prolonged absence from them. For instance, if you regularly experience a specific background noise in your home (e.g., distant traffic sounds) that carries no threat, the context of your home can become associated with that noise. When you return after a long vacation, the familiar context can “prime” the representation of that noise into its A2 state even before you explicitly hear it, causing you to habituate more quickly, or even immediately, to it compared to someone new to that environment.

---

Priming is like receiving a duplicate memo. If the memo is already in your “recently processed” outbox (A2) or even still in your “urgent” inbox (A1) because you just saw it, then a new copy of that exact same memo won’t trigger the same urgent “new information” response. It won’t go into your “urgent” inbox with the same priority. This means you’ll react to it less strongly (like habituation to a repeated sound) or learn less new information about it (like blocking a new association, because the old information is already “priming” the outcome). If the context of your office (the “background”) already makes you anticipate a certain type of memo (like retrieval-generated priming), then when that memo actually arrives, it feels less “new” or “surprising,” and you process it with less intensity.

Response to Indeterminacy in Earlier Priming Formulations

Wagner’s Sometimes Opponent Processes (SOP) model was developed to provide a more determinate and precise account of how priming influences stimulus processing and response, addressing ambiguities present in earlier formulations of priming theory. A fundamental problem with the priming theory is that the combined effect of the Conditioned Response and the Unconditioned Response sometimes resulted in more total US representation in short-term memory. This happened more frequently than when only the US was present. Priming could not definitively predict if you would see diminution or, conversely, facilitation (an increase in the UR) (Wagner, 1981, p. 42).

Basically, priming theory could not account for an increased response to a stimuli after being primed for it. Sometimes knowing an adverse event is coming creates a significant amount of distress that is then amplified when the event finally occurs. Usually the US produces a diminished response when there is a primer for the US. However, the “Sometimes Opponent” part suggests that the A2 can also contribute to the overall attention to the response.

Accordingly, the SOP theory created a mathematical representation to predict when the A2 will diminish the US response and when it will facilitate the US response.

Relationship to Opponent-Process Theories of Motivation

The “Sometimes Opponent Processes” (SOP) model derives its name partly from its formal similarities with other established “opponent-process” theories, notably proposed by Solomon and Corbit (1974). These theories share the fundamental idea that when a stimulus is presented, it triggers an initial “primary process” (referred to as the “a process” or A1 state in SOP). This primary process, in turn, automatically elicits an opposing “secondary process” (known as the “b process” or A2 state in SOP). This secondary process has a slower onset. It exhibits a gradual buildup and a more prolonged decay compared to the primary process. Its quality is hedonically opposite to that of the initial state. The “sometimes” in SOP highlights that while it accounts for these opposing effects, the theory doesn’t strictly assume that the primary and secondary processes will always contribute to a response with opposite signs.

However, SOP distinguishes itself from earlier opponent-process formulations in key ways. Crucially, SOP posits that the A1 and A2 states are inherent to the representation of any stimulus, including the conditioned stimulus (CS), and not solely the unconditioned stimulus (US). While associative learning of “A states” and “B states” is discussed in the broader opponent-process theory (Solomon & Corbit, 1974; Solomon, 1981). SOP explicitly incorporates these states into the fundamental processing of both conditioned and unconditioned stimuli. Furthermore, SOP proposes that the secondary (b or A2) process can directly mediate the observed conditioned response (CR). This means that the conditioned response might mimic the later, decaying characteristics of the unconditioned response, as both are influenced by the A2 state. In cases where the A1 and A2 states contribute to a response with opposing signs, the CR itself can appear antagonistic or compensatory to the initial unconditioned response.

Influence of Information Processing Theories

SOP draws heavily on the conceptualization of the memory system as a graph structure with interconnected representative nodes. It distinguishes between inactive (I) elements and two states of activity: a primary activity (A1s) and a secondary activity (A2s) (Wagner, 1981, p. 10).

• A1 State: Elements enter the A1 state upon stimulus presentation, and this state is directly related to the primary response to the stimulus. There are severe limitations on the number of nodes that can be in the A1 state concurrently.
• A2 State: Elements decay from the A1 state to the A2 state, which is considered a refractory state where elements are less effective in provoking A1 activity and persist for a longer period (Wagner & Brandon, 2001, p. 44). Associatively promoted nodal activity is assumed to be identical to a “decayed” version, often provoked directly into the A2 state by an associated CS, unlike a US which provokes into A1. There are less severe, but still real, limitations on the A2 state.

See Information Processing Theory for more information on this theory

Advancement Over Traditional Conditioning Models (e.g., Rescorla-Wagner)

• Quantitative Detail: SOP is noted for its sufficient quantitative detail to provide unambiguous descriptions and testable predictions for a broad spectrum of phenomena, a key advantage over theories like Groves and Thompson’s Dual Process Theory or Sokolov’s Comparator Theory (Uribe-Bahamonde et al., 2019).
• Real-time Processing: Unlike trial-level models such as the original Rescorla-Wagner model (1972), SOP is a real-time model that addresses moment-to-moment variation in stimulus processing and temporal dynamics. It explicitly considers when representative elements are active and inactive in relation to stimulus presence and absence (Wagner & Brandon, 2001).
• Explaining Habituation: SOP accurately describes all 9-10 behavioral regularities of habituation. It distinguishes between:
  • Within-session effects (short-term habituation): Explained as resulting from self-generated priming, where recent presentations of the stimulus cause its representation to be in the A2 (refractory) state, making it less effective in provoking A1 activity (Uribe-Bahamonde et al., 2019) .
  • Between-sessions effects (long-term habituation): Explained by retrieval-generated priming, where the context associated with the repeatedly presented stimulus retrieves its representation from memory into the A2 state, causing it to be “primed” and less effective upon subsequent presentations (Wagner, 1981, p. 28).
  • Contextual Influence: SOP offers specific mechanisms for how context influences stimulus processing, including “context disruption” where the presentation of an explicit stimulus transiently disturbs the processing of the context, allowing for more excitatory learning and avoiding inhibitory associations during inter-trial intervals (Uribe-Bahamonde et al., 2019).
• Broad Explanatory Scope: While focusing on habituation, SOP’s utility has been demonstrated in a variety of other associative learning phenomena, including occasion setting, timing, divergence of response measures, trial spacing, cue competition, causal learning, and mediated conditioning. It also considers how emotional responses might be represented by more delayed decay processes compared to sensory responses, explaining differential context specificity in habituation measures (Uribe-Bahamonde et al., 2019).

---

In essence, SOP is a sophisticated, quantitative, real-time model of memory processing that builds on priming and opponent-process concepts, grounded in information-processing principles. It provides a detailed, mechanistic account of how stimulus representation states (A1, A2, I) change over time and how these changes influence behavior and associative learning, particularly for phenomena like habituation and cue competition.

The Library Analogy

Imagine your brain is like a well-organized library with millions of books (stimulus elements).

• Inactive (I) state is like a book sitting quietly on the shelf.
• Primary activity (A1) state is like you actively reading a book – it’s fully occupying your attention and influencing your current thoughts and actions.
• Secondary activity (A2) state is like you’ve just finished reading the book, and it’s still lying open on your desk, readily accessible but not actively being read. While it’s there, you’re less likely to pick up a new copy of that same book to start reading from scratch.
• Habituation is like repeatedly seeing the same person. The first time, you pay full attention (A1). Each subsequent time, they’re still “on your desk” (A2), so you give them less fresh attention (less A1 activation), leading to a decline in your overt reaction.
• Self-generated priming is you repeatedly picking up the same book.
• Retrieval-generated priming (long-term habituation) is like the library’s “context” (the genre, the shelf it’s on) reminding you that you’ve seen that book before, so you don’t pick it up with fresh interest even if you haven’t seen it recently.

---

SOP, then, is like a highly detailed, real-time simulation of this library’s activity, predicting exactly which books are open, how many are on the desk, and how that affects your desire to re-read them or form new connections between books.

Empirical Evidence and Applications

SOP has been extensively tested in classical conditioning paradigms. The model accounts for a variety of challenging findings, such as:

• Spontaneous Recovery: After extinction, the conditioned response can reappear following a delay. SOP explains this by suggesting that the memory trace moves from the A2 to the I state, allowing the original association to re-emerge when the CS is re-presented (Wagner, 1981).
• Latent Inhibition: Pre-exposure to a conditioned stimulus (CS) without an unconditioned stimulus (US) retards subsequent conditioning. SOP posits that repeated CS presentations lead to the representation being more likely in the A2 state, inhibiting the formation of new associations (Lubow & Gewirtz, 1995).
• Conditioned Inhibition: SOP explains how inhibitory associations can form when the CS predicts the absence of the US, based on the memory states’ activation patterns (Denniston et al., 2001).

---

Moreover, SOP’s computational nature has made it adaptable for simulation studies and has influenced the development of connectionist models in psychology (Wagner & Brandon, 2001).

Extensions and Modifications

Researchers recognized the need to make the Sometimes Opponent Processes (SOP) model more flexible to account for complex learning phenomena. One such extension comes from Dickinson and Burke (1996), who offered a modification of Wagner’s (1981) SOP model. This revised framework, considered a “modified associative theory”, proposes that both cues that are currently present and those that are absent can undergo changes in their associative strength.

A key assumption of this modification is that the relevance of an absent cue to this change depends on its within-compound association with the cue that is present on a given trial. Crucially, for an absent cue, the change in its associative weight is in the opposite direction to the change in the association of a present cue. This modification specifically suggests that within-compound associations are essential for retrospective revaluation (learning about absent cues), but not for direct learning (learning about present cues). This distinction aligns with experimental findings showing that these associations are significantly correlated with retrospective revaluation effects but not with direct learning effects.

Comparisons with Other Models

The Sometimes Opponent Processes (SOP) model is often compared with other prominent associative models like those proposed by Pearce-Hall (1980) and Mackintosh (1975). While these alternative models largely focus on variations in the associability or effectiveness of the conditioned stimulus (CS), they differ in their specific mechanisms.

For instance, the Pearce-Hall model suggests that a CS loses its associability when its consequences are accurately predicted, meaning the stimulus is effectively processed only when it is not an accurate predictor of its consequences.

In contrast, Mackintosh’s model posits that a CS’s associability (represented by a learning-rate parameter ‘a’) increases if it predicts reinforcement more accurately than other stimuli present, and decreases if it predicts reinforcement less accurately. Both of these models represent a significant departure from earlier theories by focusing on how the CS’s processing changes with experience.

Dynamic Activation States of Memory Representations

Notably, SOP distinguishes itself by emphasizing the dynamic activation states of memory representations and their temporal properties, specifically proposing initial primary processes (A1 states) and opposing secondary processes (A2 states) that are intrinsic to both the CS and the unconditioned stimulus (US). This emphasis on time and memory states allows SOP to account for phenomena in a more mechanistic way. For example, SOP explains conditioned inhibition by the activation of the CS in its A1 state coinciding with the US in its A2 (decayed or absent) state, leading to an inhibitory association.

Furthermore, researchers have proposed several extensions to SOP to increase its flexibility and accommodate complex learning phenomena. For instance, Holland (1983) introduced modifications to address sensory preconditioning and retrospective revaluation, while Dickinson and Burke (1996) offered a revised SOP framework (sometimes called SOP 2.0) that refined assumptions about the learning process and memory trace decay to better fit empirical data.

This highlights SOP’s unique framework, which, through its detailed account of memory states and their temporal dynamics, can explain a broader range of nuanced conditioning effects beyond what other models that solely focus on changes in associability might readily address.

Critiques and Limitations

Despite its unique strengths in explaining learning through the dynamic activation states of memory representations and their temporal characteristics, the Sometimes Opponent Processes (SOP) model has faced a need for greater flexibility to account for the full complexity of learning phenomena. This necessity led researchers to propose several extensions to SOP, such as the revised framework offered by Dickinson and Burke (1996), sometimes referred to as SOP 2.0. These revisions aimed to adjust assumptions about the learning process and memory trace decay to better fit empirical data. For example, while the original SOP model predicted that the detrimental effects of conditioned stimulus (CS) preexposure and overshadowing treatments would simply add together, empirical observations revealed more nuanced interactions, driving the need for a modified theoretical approach (Denniston et al., 2001, p. 80).

Critiques to Extended Theories of SOP

However, even these extended versions of SOP have their limitations and critiques. A notable concern with Dickinson and Burke’s (1996) revised SOP model is its considerable flexibility as a function of its parameters, which can lead to predictions that are too ambiguous to allow for precise empirical testing. This inherent adaptability can make it challenging to definitively confirm or refute specific aspects of the model. Furthermore, the revised SOP has struggled to consistently account for all complex empirical findings. For instance, it failed to predict cases where the combined effects of CS preexposure and overshadowing completely cancel each other out, surprisingly eliminating any observed learning deficit.

Additionally, the model has encountered difficulty explaining situations where extinguishing a cue after training should seemingly have no effect on another cue, but instead, it either affects it paradoxically or even predicts an increase in excitatory value contrary to observations (Dennison, 2001, p. 83). These challenges indicate areas where even the modified SOP framework requires further development for a complete and consistent explanation of associative learning.

Implications and Future Directions

The Sometimes Opponent Processes (SOP) model stands out among associative learning theories by deeply focusing on the dynamic activation states of memory representations and their temporal characteristics. Unlike alternative models such as Pearce-Hall and Mackintosh, which primarily emphasize changes in attentional processes or a stimulus’s ‘associability,’ SOP uniquely posits that memory representations move through distinct primary (A1) and secondary (A2) activation states that are crucial for understanding learning.

This framework allows SOP to mechanistically explain phenomena like conditioned inhibition, where an inhibitory association forms when a conditioned stimulus’s (CS) A1 state overlaps with an unconditioned stimulus’s (US) A2 (decayed or absent) state. Furthermore, SOP has been quantitatively shown to accurately describe all 9-10 behavioral regularities of habituation, explaining both short-term (within-session) and long-term (between-session) effects through different types of ‘priming’. Recognizing the model’s need for greater flexibility to tackle more intricate learning scenarios, researchers have proposed several extensions.

A notable example is the revised SOP framework (SOP 2.0) offered by Dickinson and Burke (1996), which aimed to refine assumptions about the learning process and memory decay to better align with new empirical data, extending SOP’s utility to phenomena like causal learning and cue competition.

Ambiguous Parameters

Despite these advancements, even the extended SOP models face ongoing limitations and critiques. A significant challenge, as previously noted, lies in the inherent flexibility of its parameters, which can sometimes lead to predictions that are too ambiguous for precise empirical testing . This concern echoes broader critiques of complex theoretical models that may contain “too many degrees of freedom,” making it difficult to definitively confirm or refute specific aspects. While the sources don’t explicitly state SOP’s failure to predict phenomena like the complete cancellation of CS preexposure and overshadowing effects or paradoxical changes after extinction of a cue (as you previously described), they highlight that alternative models, such as the extended comparator hypothesis, can fully explain these counteractive outcomes, implicitly suggesting a struggle for SOP in these complex interactions.

Critiques of similar models also indicate that they may “fail to account for many of the other data” in specific contexts of overshadowing and latent inhibition, even if they explain some aspects. Moreover, some complex learning, such as certain forms of mediated conditioning, are acknowledged to be “outside the scope” of even revised associative theories like Dickinson and Burke’s work, indicating areas where the current framework may not fully apply. This ongoing need to refine its mechanisms and specify its parameters more precisely represents a key direction for future theoretical work to fully capture the nuances of associative learning.

Associated Concepts

• Classical Conditioning: This is a learning process first described by Ivan Pavlov. In this process, a neutral stimulus becomes associated with an unconditioned stimulus. Consequently, this association elicits a conditioned response.
• Automatization Theory: This theory explains how tasks become automatic through practice and repetition, impacting cognitive, motor, and social skills. The theory involves three stages: cognitive, associative, and autonomous.
• Tolman’s Rat Experiments: These experiments conducted by psychologist Edward C. Tolman revealed the rats’ latent learning and formation of cognitive maps, challenging behaviorism and expanding cognitive psychology’s understanding of internal mental processes and spatial navigation.
• Applied Behavior Analysis (ABA): This therapy style is a scientific method focused on understanding and improving human behavior using evidence-based strategies. It addresses challenges such as autism by employing techniques to reinforce positive behaviors while reducing maladaptive ones.
• Watson’s Fear Conditioning: This explains how individuals learn to associate certain stimuli with fear responses. Classical conditioning associates an initially neutral stimulus with an aversive event, leading to a learned fear response when the neutral stimulus is later encountered.
• Law of Contiguity: This refers to the concept that the mind associates two events or stimuli experienced close together in time and/or space.
• Behavior Modification: This concept rooted in behaviorism aims to shape behavior through reinforcement and punishment. Techniques like positive reinforcement, negative reinforcement, and punishment are key.
• Associative Learning: This is a type of learning where an individual or animal learns to associate two or more stimuli or events. This can involve learning that certain environmental cues predict specific outcomes or that certain actions lead to particular consequences.

A Few Words by Psychology Fanatic

In conclusion, the Sometimes Opponent Processes (SOP) Model represents a significant advancement in our understanding of associative learning by intricately weaving together concepts from priming theory and opponent-process dynamics. As we explored throughout this article, SOP offers a robust framework for examining how internal memory states influence behavior in response to environmental stimuli. By highlighting the distinct activation stages—A1 and A2—and their interplay, the model elucidates complex phenomena such as habituation, conditioned inhibition, and emotional responses that previous theories struggled to explain adequately.

As we reflect on the implications of SOP within contemporary psychological research, it is clear that its contributions extend beyond traditional conditioning paradigms. Markedly, SOP’s impact reaches further aspects. The ongoing evolution of learning theory continues to be shaped by insights derived from SOP’s nuanced approach to memory processing and behavioral responses. While challenges persist in fully capturing the complexities of learning phenomena, SOP’s legacy proves its relevance and importance. Accordingly, it advances our understanding of human behavior and cognition.

This dynamic interplay between theoretical innovation and empirical validation underscores not only the richness of psychological inquiry but also invites further exploration into how these principles can inform real-world applications across various domains.

Last Update: July 27, 2025

References:

Denniston, James C.; Savastano, Herndon I; Miller, Ralph R. (2001). The Extended Comparator Hypothesis: Learning by Contiguity, Responding by Relative Strength. In: R. R. Mowrer & S. B. Klein (Eds.), Handbook of Contemporary Learning Theories (pp. 65–118). Lawrence Erlbaum Associates. ISBN: 9780805833348; APA Record: 2000-00896-000
(Return to Main Text)

Spotlight Article:

Dickinson, A.; Burke, J. (1996). Within-compound associations mediate the retrospective revaluation of causality judgements. Quarterly Journal of Experimental Psychology, 49B(1), 60–80. DOI: 10.1080/713932614
(Return to Main Text)

Holland, P. C. (1983). Occasion setting in Pavlovian conditioning. In: M. L. Commons, R. J. Herrnstein, & A. R. Wagner (Eds.), Quantification of reinforcement (Vol. 1, pp. 183–201). Lawrence Erlbaum Associates. DOI: 10.1016/S0079-7421(08)60488-0
(Return to Main Text)

Lubow, R. E.; Gewirtz, J. C. (1995). Latent inhibition in humans: Data, theory, and implications for schizophrenia. Psychological Bulletin, 117(1), 87–103. DOI: 10.1037/0033-2909.117.1.87
(Return to Main Text)

Mackintosh, N. J. (1975). A theory of attention: Variations in the associability of stimuli with reinforcement. Psychological Review, 82(4), 276–298. DOI: 10.1037/h0076778
(Return to Main Text)

Pearce, J. M.; Hall, G. (1980). A model for Pavlovian learning: Variations in the effectiveness of conditioned but not of unconditioned stimuli. Psychological Review, 87(6), 532–552. DOI: 10.1037/0033-295X.87.6.532
(Return to Main Text)

Rescorla, R. A.; Wagner, A. R. (1972). A theory of Pavlovian conditioning: Variations in the effectiveness of reinforcement and nonreinforcement. In: A. H. Black & W. F. Prokasy (Eds.), Classical Conditioning II: Current Research and Theory (pp. 64–99). Appleton-Century-Crofts. ISBN: 9780390718013; APA Record: 1973-06348-000
(Return to Main Text)

Solomon, R. L.; Corbit, J. D. (1974). An opponent-process theory of motivation: I. Temporal dynamics of affect. Psychological Review, 81(2), 119–145. DOI: 10.1037/h0036128
(Return to Main Text)

Solomon, Richard L. (1981). The Opponent-Process Theory of Acquired Motivation. American Psychologists. DOI: 10.1037/0003-066X.35.8.691
(Return to Main Text)

Uribe-Bahamonde, Y. E.; Becerra, S. A.; Ponce, F. P.; Vogel, E. H. (2019). A quantitative account of the behavioral characteristics of habituation: The sometimes opponent processes model of stimulus processing. Frontiers in Psychology, 10, Article 504. DOI: 10.3389/fpsyg.2019.00504
(Return to Main Text)

Spotlight Book:

Wagner, A. R. (1981). SOP: A model of automatic memory processing in animal behavior. In: N. E. Spear & R. R. Miller (Eds.), Information Processing in Animals: Memory Mechanisms (pp. 5–47). Lawrence Erlbaum Associates. ISBN: 9780898591576; DOI: 10.4324/9781315798820
(Return to Main Text)

Wagner, A. R.; Brandon, S. E. (2001). A componential theory of Pavlovian conditioning. In R. R. Mowrer & S. B. Klein (Eds.), Handbook of Contemporary Learning Theories (pp. 23–64). Lawrence Erlbaum Associates. ISBN: 9780805833348; APA Record: 2000-00896-000
(Return to Main Text)