15-2: Cognitive Abilities in Nonhuman Animals

Psychology of Learning

Module 15: Comparative Cognition

Part 2: Cognitive Abilities in Nonhuman Animals

Looking Back

Part 1 established evolutionary foundations for comparative cognition—natural selection as the mechanism driving cognitive evolution, common descent linking species through shared ancestry, & the comparative approach integrating ethology, psychology, & neuroscience. Part 2 now examines specific cognitive abilities in nonhuman animals, asking what animals can do & how their abilities compare across species.

Animal Intelligence

Defining animal intelligence presents immediate challenges. Human intelligence is typically conceptualized as general cognitive ability—the capacity to reason, solve problems, learn from experience, & adapt to new situations. However, applying this concept across species raises fundamental questions: Is intelligence a single general ability (like Spearman’s g factor in humans) or a collection of specialized capacities shaped by each species’ ecological niche? Can we meaningfully compare intelligence across species with vastly different sensory systems, motor abilities, & environmental demands? These questions have no easy answers & continue to drive theoretical debates in comparative cognition.

Alternative approaches define intelligence more broadly as adaptive problem-solving—the ability to adjust behavior based on environmental demands & past experience. This ecological intelligence perspective recognizes that different species face different adaptive challenges requiring different cognitive solutions. A Clark’s nutcracker remembering thousands of seed cache locations demonstrates remarkable spatial intelligence; a dolphin coordinating with pod members to herd fish demonstrates social intelligence; an octopus escaping from a sealed jar demonstrates mechanical problem-solving; a honeybee communicating food locations through dance demonstrates symbolic communication. Each represents intelligence adapted to specific ecological demands rather than deficits relative to human cognition.

The nature versus nurture debate applies to animal intelligence just as it does to humans. To what extent are cognitive abilities innate (genetically determined) versus learned (shaped by experience)? Species differ dramatically in which cognitive abilities are innate versus learned, reflecting different evolutionary strategies. Highly specialized behaviors—bee navigation using polarized light, bird song templates in some species, salmon imprinting on natal streams—often show strong innate components requiring minimal learning. In contrast, flexible problem-solving & tool use typically require extensive learning from individual experience or social transmission from conspecifics. Understanding these differences illuminates how natural selection shapes cognitive development across diverse taxa.

Theory of Mind

Theory of mind is the ability to attribute mental states—beliefs, desires, intentions, knowledge—to oneself & others, understanding that others have perspectives different from one’s own. Theory of mind represents a sophisticated cognitive ability long considered uniquely human. Understanding that others have beliefs, desires, & knowledge different from one’s own requires representing not just the physical world but also the mental world—a capacity sometimes called mentalizing or mindreading. The question of whether nonhuman animals possess theory of mind has driven decades of comparative cognition research since Premack & Woodruff first asked “Does the chimpanzee have a theory of mind?” in 1978.

The classic false belief task illustrates theory of mind. Sally places a marble in a basket, then leaves. While Sally is away, Anne moves the marble to a box. Where will Sally look for the marble? Correctly predicting that Sally will look in the basket (where she believes it is) rather than the box (where it actually is) requires understanding that Sally’s belief differs from reality—she holds a false belief. Human children typically pass this task around age four, marking an important milestone in social cognitive development. Younger children & individuals with autism spectrum disorder often fail, predicting Sally will look in the box (where the marble actually is).

Arguments that animals demonstrate theory of mind cite deception, teaching, & social manipulation as evidence. Chimpanzees conceal food from dominant competitors, approaching food only when dominants cannot see them—suggesting understanding of what others can & cannot perceive. Scrub jays that have stolen from others’ caches later re-cache their own food when observed, but only if they themselves have been thieves—suggesting they attribute pilfering intentions to observers based on their own experience (“it takes a thief to know a thief”). Ravens appear to track what competitors have & have not witnessed during caching events, adjusting their protective behaviors accordingly. Great apes also show gaze-following & joint attention behaviors suggesting awareness of others’ attentional states.

However, arguments that animals lack theory of mind note that behaviors appearing to require mental state attribution might reflect simpler mechanisms—learned behavioral rules (“don’t approach food when dominant is looking”) rather than true understanding of others’ minds. This behavior-reading versus mind-reading debate centers on whether animals respond to observable behavioral cues or truly represent others’ unobservable mental states. Recent computational modeling research by Horschler and colleagues (2023) suggests nonhuman primates may possess intermediate representations—understanding what others see & know, but perhaps lacking the full sophistication of human theory of mind that includes representing others’ beliefs about beliefs (second-order theory of mind). Krupenye & Call’s (2019) comprehensive review concludes that some species share foundational social cognitive mechanisms with humans, while certain aspects of mentalizing may remain uniquely human. The field increasingly moves beyond asking “do animals have theory of mind?” toward examining which specific components of mentalizing exist across which species.

Concept Formation & Categorization

Concept formation is the ability to group different objects, events, or stimuli into categories based on shared features or relationships. Concepts provide cognitive economy by reducing unique stimuli to manageable categories. Without concepts, each encounter with an object would require learning from scratch—every tree would be a new entity rather than an instance of the category “tree.” Concept formation allows organisms to generalize from past experience to novel situations, a fundamental requirement for adaptive behavior in complex, ever-changing environments where no two encounters are exactly identical.

Animals demonstrate concept formation through categorization experiments. Pigeons trained to discriminate photographs containing trees from photographs without trees successfully categorize novel photographs—including tree species, orientations, lighting conditions, & contexts never encountered during training. This transfer to novel instances demonstrates true concept formation rather than memorization of specific training stimuli. Pigeons also form concepts for water (in various forms & contexts), specific individual humans (even when photographed in different clothing & poses), fish versus non-fish, & even abstract relational categories like “sameness” versus “difference.” Monkeys categorize photographs of monkeys versus other animals; honeybees categorize flowers by symmetry patterns; dolphins recognize themselves in mirrors, suggesting self-concept.

Object permanence—understanding that objects continue existing when not visible—represents another aspect of conceptual understanding studied across species. Piaget identified object permanence as a milestone in human infant cognitive development, emerging in stages during the first two years of life. Many species demonstrate object permanence: dogs, cats, primates, & corvids search for hidden objects, track invisible displacements (following an object’s trajectory even when it passes behind occluders), & maintain mental representations of objects when perceptual contact is lost. Some species show object permanence equivalent to 18-24 month old human infants.

Tool Use

Tool use—using an object to modify the environment or another object—was once considered uniquely human but is now documented across diverse species. Chimpanzees use sticks to extract termites from mounds, stones to crack nuts, & leaves as sponges to absorb water. New Caledonian crows manufacture hooked tools from twigs & pandanus leaves, modifying raw materials into functional implements—a capacity approaching human tool manufacture. Sea otters use stones as anvils to crack shellfish; Egyptian vultures drop stones on ostrich eggs; archerfish spit jets of water to knock prey from overhanging vegetation. Capuchin monkeys use stones as hammers & anvils with precision approaching early hominid tool use.

Tool use requires understanding causal relationships between actions & outcomes—recognizing that a stick can extend reach or a stone can crack a shell. Metatool use—using one tool to obtain another tool—demonstrates hierarchical planning: some crows & chimpanzees use short sticks to retrieve longer sticks needed to reach food. The distribution of tool use across species reflects ecological opportunity (availability of appropriate materials & problems to solve), cognitive capacity (understanding cause-effect relationships & physical principles), & social learning (opportunities to observe & learn from conspecifics). Tool use traditions vary between populations of the same species, suggesting cultural transmission of technological knowledge.

Memory Systems in Animals

Memory systems in animals share fundamental similarities with human memory while showing important differences reflecting ecological adaptations. Working memory limitations constrain animal cognition similarly to humans. Pigeons remember approximately four items in working memory tasks, comparable to estimates of human working memory capacity (Miller’s “magical number seven, plus or minus two” or more recent estimates of three to four chunks). Chimpanzees trained on numerical sequences demonstrate working memory for ordered items; the chimpanzee Ayumu famously outperforms humans on certain working memory tasks involving rapidly presented numerals that must be touched in ascending order after being masked.

Spatial memory provides particularly impressive animal examples demonstrating ecological specialization. Food-caching species like Clark’s nutcrackers cache up to 33,000 seeds across thousands of locations distributed over many square kilometers & recover them months later with remarkable accuracy—a feat requiring exceptional spatial memory capacity far exceeding typical laboratory demonstrations. Scatter-hoarding species (hiding single items in many locations) show better spatial memory than larder-hoarding species (storing many items in few locations), demonstrating that ecological demands shape memory specialization. The hippocampus, critical for spatial memory in mammals, is proportionally larger in food-caching species, illustrating how natural selection shapes brain structure to support cognitive specialization.

Episodic-like memory—remembering specific past events with what, where, & when information—has been demonstrated in several species. Clayton & Dickinson’s (1998) seminal research showed that Western scrub-jays remember not only where they cached food but also what type of food & how long ago, adjusting retrieval behavior based on whether perishable food (wax worms) has likely decayed versus non-perishable food (peanuts) remaining edible. When jays cached wax worms & peanuts in distinct locations, they preferentially recovered wax worms after short delays (when still fresh) but switched to peanuts after long delays (when worms had decayed). This integrated “what-where-when” memory demonstrates more than simple spatial recall—it requires binding multiple types of information into unified event representations.

The what-where-when paradigm has since been extended to rats, cuttlefish, great apes, & other corvid species. Recent research by Davies and colleagues (2024) with Eurasian jays demonstrates incidental encoding—remembering details that seemed irrelevant at encoding but can be recalled later when unexpectedly queried. Jays watched food hidden under cups marked with visual features irrelevant to finding the food, yet later recalled these incidental features when tested unexpectedly. This characteristic feature of human episodic memory—recalling incidental details of past experiences that weren’t deliberately encoded—suggests jays possess memory characteristics resembling human “mental time travel.” The term “episodic-like” acknowledges uncertainty about whether animals experience the subjective autonoetic awareness (conscious sense of mentally reliving the past) that accompanies human episodic recall.

Numerical Abilities

Numerical abilities in animals encompass multiple distinct capacities: numerosity discrimination (distinguishing quantities), ordinal understanding (recognizing “more” versus “less”), & in some cases precise counting. Many species discriminate numerosities without counting, using the approximate number system—representing quantities as analog magnitudes with increasing imprecision for larger numbers. This system, shared across vertebrates & even invertebrates like honeybees, follows Weber’s law: discrimination accuracy depends on the ratio between quantities (4 vs. 8 is as discriminable as 2 vs. 4) rather than absolute difference. Fish choose to join larger shoals; lions assess whether to attack based on relative group sizes; chimpanzees select larger food quantities.

Some animals demonstrate counting abilities exceeding approximate numerosity, showing evidence for precise enumeration of small quantities. Chimpanzees trained with Arabic numerals correctly label quantities, order numerals from smallest to largest, & perform simple arithmetic operations. Alex the African grey parrot accurately labeled quantities up to six & understood the concept of “zero” as representing absence—when asked “what color three?” while viewing two red blocks & four blue blocks, he correctly answered “none” for the color with zero blocks. Scalar timing—temporal discrimination following Weber’s law—demonstrates that animals represent time with precision proportional to duration, suggesting that time & number may share common magnitude representation systems in the brain.

The Natural Selection of Cognitive Skills

Why do particular species possess particular cognitive abilities? Natural selection provides the explanatory framework: cognitive abilities evolved because they enhanced survival & reproduction in ancestral environments. This adaptive specialization hypothesis predicts that cognitive abilities should match ecological demands—species facing similar challenges should evolve similar cognitive solutions regardless of phylogenetic relatedness (convergent evolution), while closely related species facing different challenges may diverge in cognitive abilities.

Comparative studies reveal that ecology predicts cognitive abilities better than phylogeny in some domains. Food-caching species across different families—corvids (jays, nutcrackers), parids (chickadees, tits), & rodents (squirrels)—all show enhanced spatial memory compared to non-caching relatives, with hippocampal volumes proportionally larger. Frugivores (fruit-eaters) requiring memory for widely scattered, seasonally available food sources show better spatial memory than folivores (leaf-eaters) whose food is abundant & predictable. Social species show enhanced social cognition compared to solitary relatives.

The social brain hypothesis proposes that demands of social living—tracking relationships, predicting others’ behavior, navigating coalitions & dominance hierarchies, managing reciprocal exchanges—drove brain & cognitive evolution in primates & other social species. Species living in larger, more complex social groups tend to have larger neocortex ratios & show more sophisticated social cognition. However, phylogenetic constraints also matter: closely related species share cognitive abilities even when ecological differences might predict otherwise, reflecting shared neural architecture & developmental programs inherited from common ancestors. The interplay between ecological adaptation & phylogenetic constraint shapes the remarkable diversity of cognitive abilities across the animal kingdom.

Looking Forward

Part 2 examined diverse cognitive abilities in nonhuman animals—theory of mind, concept formation, tool use, memory systems, & numerical cognition—revealing both impressive capabilities & significant variation across species shaped by natural selection matching cognitive abilities to ecological demands. Part 3 examines cognitive abilities potentially making humans unique, including language, complex tool use, cumulative culture, & the profound question of consciousness across species.