Ad Hoc Category: Dynamic Goal-Derived Concepts

Human cognition possesses a remarkable capacity to dynamically construct novel conceptual groupings tailored to immediate behavioral contexts, bypassing stable taxonomic structures stored in long-term memory. When confronted with idiosyncratic circumstances—such as evacuating a burning house or surviving an unexpected blizzard—individuals spontaneously synthesize functional concepts like “things to take out of a burning house” or “objects that can serve as an improvised snow shovel.” Known within cognitive psychology as an ad hoc category, this construct demonstrates that human categorization is not merely a static retrieval of pre-existing taxonomies, but an active, goal-driven computational process deeply embedded in situational demands.

Ad Hoc Category

1. Concise Definition

An ad hoc category is a specialized, novel conceptual category formed dynamically on the spot to achieve a specific goal within a particular context, rather than retrieved as an intact, pre-existing structure from long-term memory. Coined and rigorously operationalized by cognitive psychologist Lawrence Barsalou, these categories emerge spontaneously when situational demands require grouping diverse entities that share little to no intrinsic perceptual or taxonomic overlap, yet collectively fulfill a pragmatic function.

Unlike common taxonomic categories (such as “mammals,” “furniture,” or “vehicles”), ad hoc categories typically lack entrenched lexical labels, conventional linguistic representation, and fixed memory traces prior to instantiation. Despite their spontaneous assembly, ad hoc categories exhibit the standard psychometric hallmarks of natural categories, including well-defined graded structures, robust prototype effects, and consensus across individuals regarding category membership and exemplar typicality.

2. Etymology & Linguistic Origin

The term derives from the Latin phrase ad hoc, literally translating to “for this” or “for this specific purpose.” Historically utilized in legal, administrative, and philosophical discourse to describe solutions, committees, or arguments tailored strictly to an immediate problem without broader generalization, the phrase entered the lexicon of cognitive psychology through seminal empirical investigations conducted by Lawrence W. Barsalou in the early 1980s.

Barsalou integrated the Latin prepositional phrase with the psychological construct of “category” (originating from the Ancient Greek kat&emdash;goria, denoting an accusation, predicate, or fundamental ontological class) to challenge the prevailing view that concepts exist purely as static, pre-compiled semantic nodes. By pairing “ad hoc” with “category,” Barsalou encapsulated the paradox of conceptual representations that are both ephemeral in origin yet systematic and rule-governed in cognitive architecture.

3. Pronunciation & Grammatical Form

Pronunciation: Phonetically transcribed in the International Phonetic Alphabet (IPA) as /æd ˈhɒk ˈkætɪɡəri/ (British English) or /æd ˈhɑːk ˈkætəɡɔːri/ (American English).

Grammatical Form: Compound noun phrase. The constituent “ad hoc” functions as a pre-modifying attributive adjective modifying the head noun “category.” The plural form is ad hoc categories. In academic discourse, the hyphenated form ad-hoc category is occasionally observed, though standard cognitive science conventions maintain the unhyphenated Latin rendering “ad hoc category.”

4. Detailed Conceptual Explanation

To comprehend the cognitive architecture of an ad hoc category, one must contrast it with common taxonomic categories. Traditional theories of semantic memory long posited that conceptual organization reflects objective environmental regularities. For instance, birds share feathers, wings, beaks, and metabolic characteristics; these attributes form correlated feature bundles that semantic memory encodes into stable mental representations. When encountering a robin, an agent matches its features against the stored prototype or exemplar set of “bird.”

In contrast, ad hoc categories—such as “things to pack for a camping trip in the desert” or “activities to do while waiting for a delayed flight”—do not reflect correlated perceptual clusters in the physical world. A compass, sunscreen, a thermal blanket, and iodine tablets possess virtually zero physical similarity, share few perceptual primitives, and occupy distinct branches of traditional ontological taxonomies. What unites them is entirely internal: the psychological goal of the agent. The category is synthesized in working memory through the interaction of situational constraints and intentional objectives.

Critically, Barsalou demonstrated that once an ad hoc category is generated, it behaves cognitively in the exact same manner as deeply entrenched taxonomic categories. Most notably, ad hoc categories possess graded structure: members are not treated as equally representative of the category, but instead exhibit strong typicality gradients. For the ad hoc category “things to take from a burning house,” experimental participants consistently judge children, pets, and irreplaceable family heirlooms as highly prototypical exemplars, whereas a television or a generic coffee mug are considered marginal, peripheral exemplars.

This graded structure is organized around an ideal rather than a statistical average. In natural categories, typicality is largely determined by central tendency—the degree to which an exemplar possesses features common to other category members and distinct from non-members (family resemblance). In ad hoc categories, typicality is determined by zero-point proximity to an ideal value that optimizes goal fulfillment. For example, in the ad hoc category “things that could be stepped on to reach a high shelf,” the prototype is not an average of potential objects, but an object possessing the ideal properties of structural rigidity, non-slip texture, and appropriate elevation.

Furthermore, ad hoc categories illustrate the remarkable plasticity of human semantic memory. Rather than serving as an inflexible warehouse of static concepts, semantic memory operates as a generative conceptual engine capable of dynamically recombining world knowledge, causal models, and motivational states to produce on-demand cognitive representations.

5. Historical Development

The emergence of the ad hoc category concept in the 1980s marked a watershed moment in the history of cognitive science, directly challenging the foundational assumptions of the classical and early post-classical categorization paradigms:

• Classical View (Aristotelian Paradigm): Dominated philosophy and early cognitive models, asserting that categories are defined by singly necessary and jointly sufficient features. This view could not account for typicality effects, fuzzy boundaries, or flexible contextual groupings.
• The Prototype Revolution (1970s): Led by Eleanor Rosch and colleagues, prototype theory demonstrated that natural categories have continuous, graded structures organized around central tendencies. However, prototype theory initially presumed that categories were relatively static structures based on correlated perceptual features found in the environment.
• Barsalou’s Seminal Breakthrough (1983): In his groundbreaking paper “Ad hoc categories” published in Memory & Cognition, Lawrence W. Barsalou challenged the static view. He presented empirical evidence that people frequently create goal-derived, ad hoc groupings that exhibit typicality gradients without prior instantiation or established perceptual feature correlations.
• Expansion to Goal-Derived Categories (1985–1991): Barsalou expanded the framework by distinguishing purely transient ad hoc categories from stabilized goal-derived categories. Over repeated instantiations, frequently utilized ad hoc categories (such as “birthday presents for a spouse” or “diet foods”) transition into entrenched semantic memory, demonstrating the evolutionary pathway from transient mental constructions to crystallized cultural concepts.
• Grounded and Embodied Cognition Era (Late 1990s–Present): As cognitive science embraced embodied and grounded cognition—pioneered significantly by Barsalou himself—ad hoc categorization became a central pillar supporting the view that concepts are dynamic modal simulations constructed to support situated action.

6. Theoretical Foundations

The concept of ad hoc categories rests upon several interconnected theoretical frameworks across cognitive science, psycholinguistics, and philosophy of mind.

First and foremost is Situated Cognition and Perceptual Symbol Systems. Under Barsalou’s perceptual symbol system hypothesis, concepts are not amodal symbols stored in a mental dictionary. Instead, they are dynamic neural re-enactments (simulations) of sensory-motor and introspective states. Ad hoc category formation represents situated simulation in its purest form: an agent constructs a temporary conceptual simulation by assembling components across diverse sensory-motor networks to guide action within an immediate, situated environment.

Second, ad hoc categorization relies upon Theory-Based Categorization (championed by researchers such as Douglas Medin, Gregory Murphy, and Edward Wisniewski). This framework posits that concepts are structured not merely by similarity matrices or feature co-occurrences, but by intuitive, causal theories of how the world works. An ad hoc category like “objects that can serve as a doorstop” is not governed by surface similarity, but by an underlying mechanical theory linking mass, friction, gravity, and object clearance.

Third, Working Memory and Executive Control Frameworks explain the computational mechanics of ad hoc categorization. Unlike taxonomic categorization, which can occur via rapid, automatic spreading activation in semantic memory, ad hoc categorization requires active maintenance in working memory, selective attention, inhibitory control (suppressing irrelevant intrinsic attributes of objects), and relational reasoning.

7. Key Components, Types & Dimensions

Understanding the architecture of ad hoc categories requires breaking down their essential components, developmental trajectories, and functional dimensions:

• Goal-Dependency: The primary dimension defining the category. The category exists solely as a function of an underlying plan, problem, or objective. The goal dictates the inclusion criteria, evaluation metrics, and hierarchical structure of the members.
• Ideal-Based Graded Structure: Category organization is determined by how closely an exemplar approximates an ideal standard of goal fulfillment, rather than the arithmetic average of category members.
• Low Category-to-Attribute Predictability: Unlike natural taxonomies where knowing an item belongs to a category (e.g., “bird”) permits robust inductive inferences about its attributes (e.g., “has wings,” “lays eggs”), knowing an item is in an ad hoc category (e.g., “things to sell at a garage sale”) permits few inferences about its physical morphology.
• Transient vs. Institutionalized Ad Hoc Categories: Pure ad hoc categories are assembled for a singular, idiosyncratic episode (e.g., “things to throw at an aggressive goose”). If an ad hoc category is repeatedly engaged across time or social groups, it undergoes conventionalization, transitioning into an institutionalized goal-derived category (e.g., “carry-on luggage” or “camping gear”).
• Relational and Extrinsic Features: Membership in an ad hoc category is predominantly governed by extrinsic relations (how the object relates to human agents and external circumstances) rather than intrinsic features (the physical, genetic, or molecular composition of the entity).

8. Examples & Illustrative Cases

Ad hoc categories manifest continuously throughout human problem solving, clinical behavior, and strategic decision-making. The following cases illustrate their cognitive dynamics:

Case 1: Emergency Evacuation (“Things to save from a burning home”)
When an individual realizes their residence is on fire, cognitive processing immediately constructs an ad hoc category. The exemplars include family members, pets, photo albums, legal documents, and digital hard drives. These items share virtually no physical or ontological overlap. Prototypicality is dictated by the ideal: maximum sentimental or legal irreversibility coupled with portability. Items like large furniture or kitchen appliances, despite their monetary value, are immediately excluded due to situational constraints (lack of time and physical transport capacity).

Case 2: Improvised Tool Use (“Objects that can hold a loose tent flap in high winds”)
A camper encountering broken tent pegs generates an ad hoc category to prevent structural collapse. Potential members include heavy flat stones, water-filled coolers, backpacks loaded with canned food, and vehicle tire rims. The underlying cognitive process abstracts the functional requirement (“downward gravitational force exceeding the upward aerodynamic lift of the wind”) and maps it onto surrounding environmental objects, completely disregarding their conventional taxonomic designations.

Case 3: Dietary Restriction in Social Settings (“Foods I can safely consume at this wedding banquet”)
An individual diagnosed with severe celiac disease constructs an ad hoc category upon approaching a buffet table. The category pulls together items such as raw crudités, unmarinated grilled chicken breast, and packaged sparkling water, while strictly excluding wheat-flour pastries, soy sauce-glazed meats, and beer. The category is assembled dynamically through causal knowledge regarding culinary contamination risks.

9. Measurement & Assessment

Investigating ad hoc categories within experimental cognitive psychology requires distinct methodological paradigms designed to capture dynamic, real-time concept formation:

Typicality Rating Tasks: Researchers present participants with an ad hoc category frame (e.g., “ways to escape a boring conversation”) alongside an array of candidate exemplars. Participants rate each item on a Likert scale (typically 1 to 7) assessing category typicality. Statistical analyses measure inter-subject agreement to establish whether the ad hoc category demonstrates robust graded structure despite its novel nature.

Generation and Production Protocols: Participants are allotted a fixed time window (e.g., 60 seconds) to generate as many exemplars as possible for an ad hoc category. Frequency of generation, serial order of retrieval, and latency to first exemplar provide empirical metrics of cognitive accessibility, working memory retrieval dynamics, and the activation strength of relational networks.

Verification Reaction Time (RT) Paradigms: Using computerized psychological testing platforms, subjects are primed with an ad hoc context and subsequently presented with category-exemplar pairs. Response latency for affirming or denying membership is recorded. In classic experiments, typical exemplars demonstrate significantly faster verification times than atypical exemplars, confirming that prototype effects operate in spontaneously synthesized categories just as they do in stable natural kinds.

Neuroimaging Correlates: Functional Magnetic Resonance Imaging (fMRI) and event-related potential (ERP) paradigms assess the differential neural activation patterns between taxonomic retrieval and ad hoc category assembly. Ad hoc categorization consistently engages enhanced bilateral prefrontal cortex (PFC) activation, anterior cingulate cortex (ACC) engagement for conflict monitoring, and posterior parietal regions associated with selective working memory manipulation, contrasting with the dominant anterior temporal lobe (ATL) hub activation seen in standard taxonomic retrieval.

10. Applications & Practical Significance

The construct of the ad hoc category yields profound implications across multiple practical domains:

Human-Computer Interaction and Information Architecture: Traditional user interfaces organized data hierarchically along taxonomic lines (e.g., filing systems organized by file format or chronological folder). Modern user-centric design frequently embraces ad hoc categorizations, such as dynamic tagging, smart folders, and goal-oriented search filters (e.g., “files worked on while traveling last week” or “budget software for first-time homebuyers”).

Marketing and Retail Merchandising: Retail strategists frequently discard traditional aisle taxonomies in favor of ad hoc groupings that mirror consumer behavioral objectives. Supermarkets create seasonal displays combining charcoal, marinade, paper plates, bug repellent, and skewered meats under the goal-derived ad hoc category of “backyard barbecue essentials.” This lowers cognitive search costs for consumers and triggers impulse purchasing by aligning retail topography with mental simulations.

Cognitive Rehabilitation and Neuropsychology: Neurological damage, such as traumatic brain injury (TBI) or executive dysfunction secondary to prefrontal lesions, often spares basic taxonomic naming while severely impairing ad hoc category formation. Assessing a patient’s ability to generate ad hoc groupings provides clinicians with a sensitive diagnostic measure of real-world problem-solving, cognitive flexibility, and executive functioning.

Artificial Intelligence and Large Language Models (LLMs): Early rule-based AI systems failed catastrophically at flexible categorization because their ontologies were statically coded. Modern deep learning architectures and LLMs excel precisely because transformer attention mechanisms enable dynamic, context-conditioned representations analogous to human ad hoc categorization, grouping disparate entities based on prompt-specified constraints.

11. Research & Empirical Evidence

Barsalou’s initial 1983 investigations established the foundational empirical benchmarks for ad hoc categories. In a series of experiments, he presented participants with ad hoc categories (such as “things to avoid when on a diet” and “ways to kill a cockroach”) alongside common taxonomic categories (e.g., “fruit,” “vehicles”). Across four rigorous experiments, Barsalou found that:

• Participants demonstrated high intra-individual and inter-individual agreement on typicality judgments for ad hoc categories, yielding correlations nearly identical to those observed for common taxonomic categories.
• Ad hoc categories possess graded structures that reliably predict verification reaction times.
• Unlike common taxonomic categories, where typicality is determined by central tendency (maximizing similarity to category members while minimizing similarity to contrasting categories), typicality in ad hoc categories was overwhelmingly predicted by fidelity to ideals (the dimension that maximally satisfies the category goal).

Subsequent empirical work by Barsalou and Sewell (1985) examined the perspective-taking plasticity of these categories. They demonstrated that participants could reliably adopt alternate personas (e.g., rating “things to take on a weekend trip” from the perspective of an elderly scholar versus an athletic college student) and rapidly reconfigure the graded structures of ad hoc categories, proving that these structures are dynamically synthesized in working memory rather than retrieved as immutable conceptual templates.

Further investigations by Gregory Murphy (2002) and Edward Wisniewski (1997) highlighted that ad hoc category processing relies heavily on explanatory and causal knowledge rather than mere linguistic association. Neurocognitive studies by Ross and Spalding (1994) confirmed that categories constructed during active problem solving leave unique episodic traces that gradually reorganize background semantic networks, providing the empirical mechanism by which transient ad hoc concepts stabilize into entrenched semantic structures over time.

12. Cultural & Cross-Cultural Considerations

While the computational mechanism of forming ad hoc categories appears universal across all human populations, the content, ideals, and contextual constraints governing their generation are heavily shaped by cultural norms, ecological niches, and socio-economic realities.

In cross-cultural cognitive research, individuals from agrarian or subsistence societies frequently group items based on dynamic ecological and functional relationships rather than Western-style abstract taxonomic categories. Anthropological psychologists have noted that what Western standardized intelligence tests historically classified as “failure” to categorize taxonomically (e.g., grouping a hoe with a potato rather than grouping a hoe with a shovel) actually reflects an advanced reliance on functional and situational categorization akin to ad hoc categorization.

Moreover, the ideals that anchor the graded structure of an ad hoc category vary dramatically between cultural groups. For an ad hoc category like “appropriate gifts for an elder,” a collectivist culture may elevate exemplars characterized by respect, formal utility, and communal status, whereas an individualist society might prioritize personalization, emotional intimacy, or financial value. The cognitive architecture—relying on ideals rather than central tendency—remains identical, but the culturally transmitted ideal radically alters the composition of the prototype.

13. Criticisms, Debates & Limitations

Despite its widespread acceptance, the construct of the ad hoc category has sparked significant theoretical debate within cognitive psychology and philosophy of mind:

• The Boundary Problem: Critics have questioned whether a bright conceptual line truly separates ad hoc categories from common taxonomic categories. Some theorists argue that all categorization is to some degree situated and goal-dependent, suggesting that “ad hocness” is a continuous spectrum of memory familiarity rather than a distinct categorical taxon.
• Cognitive Load and Tractability: Proponents of traditional modular theories argue that dynamic ad hoc categorization imposes excessive computational overhead on working memory. They contend that the mind must possess pre-compiled associative heuristics to constrain the limitless possibilities of ad hoc synthesis, lest the agent suffer from combinatorial explosion when solving simple behavioral tasks.
• Stability and Consensus Discrepancies: While Barsalou demonstrated inter-subject agreement on typicality, several subsequent studies noted that between-subject agreement on the initial generation of exemplars for ad hoc categories is markedly lower than for established taxonomic categories. People agree on what makes a good exemplar once presented, but their spontaneous generation pathways are far more idiosyncratic.
• Ontological Status in Philosophy of Mind: Some conceptual essentialists question whether ad hoc categories constitute “genuine” concepts at all, viewing them instead as temporary propositional descriptions or complex linguistic predications rather than unified cognitive representations.

14. Related Terms & Distinctions

To ensure conceptual clarity, ad hoc categories must be distinguished from several closely aligned constructs in the psychological literature:

• Taxonomic Category: A category organized around natural correlated feature structures and biological or ontological classification (e.g., “reptile,” “mineral”). Unlike ad hoc categories, taxonomic categories are stable, highly entrenched in long-term memory, and have high category-to-attribute predictability.
• Goal-Derived Category: A broader overarching class of categories organized around goal fulfillment. While all ad hoc categories are goal-derived, not all goal-derived categories are ad hoc. Established goal-derived categories (e.g., “diet foods,” “exercise equipment”) have become permanent, institutionalized fixtures of semantic memory through repeated use, whereas ad hoc categories are synthesized on the fly.
• Relational Category: A category defined by a core relation or relational schema rather than intrinsic features (e.g., “barrier,” “fuel,” “predator”). Relational categories can be deeply entrenched and institutionalized, whereas ad hoc categories are typically constructed for an immediate, non-entrenched operational context.
• Schema / Script: Organized knowledge frameworks describing sequences of events or spatial configurations (e.g., “dining at a restaurant script”). Schemas provide the background contextual knowledge from which ad hoc categories are extracted, but a schema is an entire episodic framework, whereas an ad hoc category is a selective conceptual grouping of distinct items.
• Folk Taxonomy: Culturally transmitted, everyday vernacular classifications of the natural world (e.g., “bugs,” “weeds”). Folk taxonomies are stable, shared linguistic traditions, whereas ad hoc categories can be wholly novel and unique to an individual’s immediate predicament.

15. Summary / Key Takeaways

The ad hoc category exemplifies the supreme flexibility, context-sensitivity, and goal-directed nature of human cognition. Key takeaways regarding this construct include:

• Ad hoc categories are novel groupings constructed dynamically in working memory to serve an immediate goal within a specific contextual environment.
• They do not require prior long-term memory instantiation or shared perceptual features among their exemplars.
• Despite their spontaneous nature, they exhibit classical prototype phenomena, possessing reliable graded structures and typicality effects.
• Unlike taxonomic categories, whose prototypes reflect the central tendency of correlated features, ad hoc categories are structured around proximity to an ideal that optimizes goal satisfaction.
• Through repeated usage over time, transient ad hoc categories can institutionalize into permanent, goal-derived semantic categories.
• They serve as foundational evidence for grounded, situated, and theory-based models of human conceptual architecture.

References

• Barsalou, L. W. (1983). Ad hoc categories. Memory & Cognition, 11(3), 211–227. https://doi.org/10.3758/BF03196968
• Barsalou, L. W. (1985). Ideals, central tendency, and frequency of instantiation as determinants of graded structure in categories. Journal of Experimental Psychology: Learning, Memory, and Cognition, 11(4), 629–654. https://doi.org/10.1037/0278-7393.11.4.629
• Barsalou, L. W. (1991). Deriving categories to achieve goals. In G. H. Bower (Ed.), The Psychology of Learning and Motivation (Vol. 27, pp. 1–64). Academic Press. https://doi.org/10.1016/S0079-7421(08)60122-X
• Barsalou, L. W., & Sewell, D. R. (1985). Contrasting the representation of scripts and categories (Technical Report No. 10). Emory Cognition Project.
• Murphy, G. L. (2002). The Big Book of Concepts. MIT Press. https://doi.org/10.7551/mitpress/1602.001.0001

Ultimately, the discovery and operationalization of ad hoc categories transformed the landscape of cognitive science. By proving that human conceptual structures are not restricted to rigid, pre-cataloged inventories of sensory properties, cognitive psychology revealed that categorization is an intrinsically creative, adaptable process. Whether securing safety during an emergency or generating novel computational solutions, the human ability to spontaneously construct ad hoc categories underscores the vital synergy between motivation, environment, and higher-order reasoning.