Lecture 7 - Decision Making

• Psychology 2210B is covering Decision Making
• Instructor: Krista Macpherson, PhD
• Date: March 3rd, 2025
• Exam 2 will be next week and in the same format as Exam 1

Overview

• Chapter 7 focuses on Decision Making, including free will, foraging theories, laboratory studies, and Behavioural Economics.
• Chapter 7 also includes Risky Choice, examining gambling/suboptimal choice, delay discounting, individual differences, and the Somatic Marker hypothesis.
• Chapter 8 content is conditional based on available time, and only covered material will be on the exam.

Win/Shift Task on the Radial Maze: Comparative Performance

• Rats, pigeons and dogs were tested to measure correct choices in first four arm selections in a Win/Shift Task.

Importance of the Win/Shift Task

• Shows an animal's tendency to seek new food sources rather than return to depleted arms.
• Win/shift tasks relate to spatial memory and are a component of optimal foraging.

Decision Making Defined

• Decision making constitutes the process of selecting one course of action from a variety of options.
• Choice is the outcome of this decision-making.
• A motor action or verbal response indicates a choice.
• Foraging by Northwestern crows exemplifies decision making.

Free Will Considerations

• Restriction on free will violates human rights like freedom of speech and academic freedom across cultures.
• Philosophers from Hindu, Buddhist, Islamic, and Western traditions have questioned freedom of thought, especially with unconscious acts.
• Brain imaging studies challenge the existence of free will.
• Decisions occur up to 10 seconds before conscious awareness.

Foraging Theories

• Optimal Foraging
• Marginal Value Theorem
• Ideal Free Distribution Model
• Collective decision making is relevant for coordinated activities like migration but is not necessarily a foraging theory.

Optimal Foraging Theory

• The idea that decision making evolved to maximize fitness is often observed in foraging behaviors.
• The energy acquired during food consumption offsets the energy expended to obtain it.
• Optimal Foraging Theory states that organisms forage in a way that maximizes net energy gain per unit of time.
• Chipmunks are central place gatherers that demonstrate this.

Factors Impacting Optimal Foraging Theory

• Travel Time
• Handling
• Predation
• A key theme is that animals must have a net surplus of calories after factoring out "operating expenses".
• If travel time increases, the number of worms a bird carries also increases.

Travel Time

• Transport becomes more difficult with heavier loads.
• Animals carry more food on longer trips to make it "worth it".
• The principle applies to humans: convenience stores for few items, grocery stores for many.

Kacelnik Study (1984)

• Birds were trained to retrieve worms from a feeder at increasing distances from their nest.
• As the bird needs to travel farther, it increases the number of worms it carries.

Handling Time

• Handling is the time and energy needed to extract consumable food from its source.
• Shore crabs feeding on blue mussels demonstrate handling by breaking open shells.
• Handling is the quantity of flesh extracted per amount of handling time.
• Selection of intermediate-sized mussels maximizes resource capture (Elner & Hughes, 1978)

The Imperfections of Handling Time

• Animals can make errors due to other cognitive processes.
• Wasting time searching for intermediate shells even if they are scarce and larger shells are abundant occurs.

Handling Time in Crows

• Zach (1979) observed crows feeding on whelks, selecting only larger whelks (3.5-4.4cm).
• They fly about 5m high to drop whelks.
• Multiple attempts are sometimes made to break open a single whelk.
• Northwest Crows dropping large whelks picking them up maximizes calories. Dropping from 5m minimizes time/energy expended in handling

Optimality in Whelk Breaking

• Large whelks are more likely to shatter at 5m.
• Drops higher than 5m do not greatly improve breakage.
• Drops lower than 5m reduce breakage.

Challenges in Optimality Models

• Creating an optimality model can be difficult because birds are choosing smaller mussels than would be predicted based on caloric value/handling time in Eurasian Oystercatchers.

Optimal Foraging Models

• Model A considers the Caloric value of mussel/handling time.
• Model B includes large mussels that birds cannot open.

Predation Impact

• Animals are most vulnerable when away from their home base searching for food.
• Optimal foraging is thus compromised in the presence of high predation.
• Leafcutter ants forage at night to avoid predator flies (Orr, 1992).
• Rat foraging bouts are inhibited if predators are detected (Blanchard et al., 1992).

Landscape of Fear

• In Yellowstone National Park in 1995, reintroduced wolves caused elk to increase vigilance by 50%, hiding in wooded areas rather than grazing openly.

Marginal Value Theorem

• Food is not evenly distributed; it occurs in patches.
• As patch resources deplete, fitness benefits of staying decline.
• Elk Foraging Patches was studied by Van Moorter et al. (2009)

Marginal Value Defined

• Marginal value is the point where net energy gain is lower than surrounding areas.
• The decision to move depends on amount of resources, distance between patches, and time spent at a patch.

Cowie's Study (1977)

• Birds were trained to forage in artificial patches (cups filled with sand and prey in an aviary.)
• As travel distance between patches increases, birds spend more time at a patch.

Lima's Study (1984)

• This study set up three habitats with empty logs filled with different numbers of prey items.
• Each habitat had poor and rich patches.
• Rich patches in the 1st habitat had 24 items.
• The 2nd had 12.
• The 3rd had 6.
• Birds foraged these patches for several days, and these birds developed diverse strategies per patch.

Lima's Findings (1984)

• Birds abandon holes immediately if they do not find food in rich prey patches.
• Birds invest more time foraging in prey patches which have been depleted.

Ideal Free Distribution Model

• Many species forage in groups.
• Payoff decreases if animals congregate at one patch.
• Distributing amongst patches is better than all staying in one rich patch.
• Animals may not be "free" to move because of social or physical constraints, demonstrating that it is not a perfect system.
• The basic idea is that the number of animals at a food patch is proportional to the amount of food available.

Milniski's Study (1979)

• In stickleback fish, both ideal free distribution model characteristics were recorded by and food dropped into aquarium.
• Fish aggregating in each end recorded.
• When food is delivered in a 5:1 ratio stickleback fish aggregate accordingly.

Nash Equilibrium

• A theory describes forces governing group dynamics.
• The decision-making of others in a group must be considered because individual decisions cannot be understood in isolation.
• Players are in Nash equilibrium if they are making the best decision, considering others' decisions, as long as those decisions remain unchanged.

Collective Decision Making

• Evolved coordinated behaviors that provide a fitness advantage to a group are characteristics of collective decision making.
• This is seen in migrating flocks of birds and marching colonies of ants.
• Collective decision making tends to be regular and predictable, governed by general principles.
• Collective decision making is robust, and a single defector will not alter group behavior.
• Collective decision making is different from cooperative behavior.

Examples of Collective Decision Making

• Stickleback fish ignore a single neighbor's movement and instead follow once a critical mass moves in one direction (Ward et al., 2012).
• Increasing group size increases the accuracy of collective decisions (Sumpter, 2006).
• Making is probably more effective/efficient than individual choice.

Confusion Effect

• Murmurations are often beautiful, but can be destructive.
• Starlings are an invasive species.
• Eugene Schieffelin released 60 common starlings in 1890 into New York's Central Park.

Beyond Foraging

• Decision making is often discussed in foraging behaviors in non-human animals but has important implications for aggression, habitat selection, and reproduction.
• Decision making is thought to be based on the lowest energy expenditure for the highest fitness benefit.
• Behavior traits in foraging may be inherited.
• The result of an individual’s interaction with the environment.
• These behaviors can be affected by cognitive processes like time, memory, and discrimination.

Choice Experiments in the Lab

• These can be conducted using concurrent schedules of reinforcement which encompasses two or more schedules in effect at the same time.
• If a pigeon has an FR5 schedule for one button and an FR2 schedule for another, it will almost exclusively hit the FR2 button.

Schedules of Reinforcement

• Schedules of reinforcement are not always that simple.
• With buttons on FI-30s and FI-45s schedules, pigeons maximizes reward by moving back and forth between the keys

The Matching Law

• The Matching Law reflects how maximal payoff occurs when animals divide time between behavioral options.
• Herrnstein (1961) stated the relative rate of responding must match the relative rate of reinforcement.
• Animals generally maximize gains across sessions.
  • The formula is BA/(BA + BB) = rA/(rA + rB).

Shortcomings of the Matching Law

• The Matching Law is not a perfect system
• Inconsistencies occurs as different amounts of effort are required, rewards vary in quality or quantity, rewards and reinforcers are delayed
• Deviations from matching law include under/overmatching, and bias.
• Bias refers to when one is leaning toward one alternative or the other, regardless of payoff.
• Undermatching is a response rate higher than expected to a lower payoff option.
• Overmatching is a response rate higher than expected to a higher payoff.

Behavioural Economics

• Concurrent choice study in animals parallels the study of economic choice by humans.
• Behavioural Economics analyzes choice behaviour using economic principles (Hursh, 1980).
• Economists explain purchasing behavior via cost/demand: rising prices lead to decreased purchasing, a phenomenon shown in animals.

Purchasing Habits of Humans

• Purchasing of necessities changes little when prices increase.
• Purchasing of non-essential items is sensitive to increased costs.
• Food is thus inelastic because it is essential and purchasing is changed little by price.
• Alcohol is elastic because it is non-essential and purchasing is sensitive to price.
• Drug and alcohol can move from elastic to inelastic for the addicted, with important Implications for addiction behaviour

Risky Choice

• Risky choice falls under decisions made under uncertainty.
• These are studied as gambling or suboptimal choice tasks.
• Evolutionary and foraging theories suggest that non-human animals should not engage in risky/suboptimal choice.

Heuristics

• Heuristics help humans and animals deal with choices in the face of uncertainty.
• These are strategies for solving problems that ignore a portion of information.
• Often referred to as "mental shortcuts" or cognitive "rules of thumb."
• May be used in the face of complex problems or incomplete information.
• Heuristics often result in a correct solutions/desired outcome.
• Often contrasted with algorithms.
• Algorithms are specific, correct instructions or procedures to lead to a correct solution. - Always results in the desired outcome - More costly in time/energy

Tversky & Kahneman's Prospect Theory

• Describes the way people choose between alternatives with uncertain probabilities, involving risk.
• States humans are bad "intuitive statisticians" because they make decisions based the potential value of losses and gains rather than the final outcome.
• The theory states evaluations of these losses and gains involves heuristics.

Sunk Cost Fallacy

• Cognitive heuristics lead to biases/systematic errors, with one being the sunk cost fallacy.
• The sunk cost is one that cannot be recovered, whether that is money spent or energy expended.
• Sunk costs should not factor into future payoffs, yet humans make this mistake, known as the Concorde fallacy.
• Loss aversion is used as an evolutionary explanation
• It has been attributed to cognitive dissonance (Festinger, 1957).
• It can be demonstrated in non-human animals.
• European Starlings prefer stimuli linked to high work output (Kacelnik & Marsh, 2002).
• Pigeons persist on one option even when another with higher payoff is introduced (Macaskill & Hackenberg, 2012).
• The same evolutionary theory applies to sunk cost fallacy to animals.
• Animals stay on a "paying off" patch avoiding the risk of a new patch.
• There is more safety in a "sure thing".

Heuristics in Animals

• Peahens prefer to mate with Peacocks that have the most spots (Petrie & Halliday, 1994).
• Some birds escape predators when they see at least two other birds fleeing (Lima, 1994).

Risk Taking

• There is uncertainty in decision making when one or more options risk is considered.
• Animals may forage in a low, consistent density patch instead of a high, unpredictable one.
• An example of a concurrent Task is a choice between 2 pellets 100% of the time and 8 pellets 25% of the time.

Risk Schedule Setups

• The low-payoff option is advantageous in the long run, but both pigeons and humans prefer the risky option (Zentall & Stagner, 2011).
• This is more pronounced in humans who gamble (Molet et al. 2012).

Motivations Behind Risky Behavior

• Organisms engage in risky behavior when the safe option cannot meet fitness needs, like starvation.
• Another possibility is that high outcomes which rarely occur are more memorable, and are easier to remember.
• Thus, individuals mistakenly think that they occur more often, known as an availability heuristic (Tversky & Kahneman, 1974).
• Risk taking affects the behavior of rats because it increases the chance of predation.

Delay Discounting

• Humans and animals prefer an immediate payoff, even if smaller than a delayed one.
• Future rewards tend to decline in perceived value.
• The longer the delay, the more the reward is discounted, but the larger the reward, the more the delay is necessary to discount it.
• Delay discounting measures impulsivity (small immediate rewards) and self-control (large delayed rewards).

Delay Discounting Paradox

• From an evolutionary perspective, the payoff should be optimized by choosing the delayed reward.
• The most explanation is delayed rewards are secured immediately because its unpredicatable.

Kacelnik-Context in Food Encounters

• Food in the wild is encountered sequentially, not simultaneously.
• Animals assign value to food via context.
• Food in a resource-sparse area is valued more than the same food in a resource-rich area.

Individual Differences

• Within-species decision making differences exist, especially with uncertain choices.
• Birds and mammals differ in preference for risky/non-risky options and small immediate vs. long delayed rewards.
• High human impulsivity leads to conditions like drug addiction (Kirby and Petry, 2004).
• Lack of prefrontal cortex maturation can result in impulsivity/risk-taking in young humans/animals.
• PFC is linked to inhibitory control/long-term planning.

Emotional Decision Making

• Emotions are described as forces that require reason to guide/tame them, potentially distort making decisions, dating back to Plato.
• Emotional decision making was formally studied until late 20th century.
• Neurologist Damasio found high scoring patients on psychological tests who still had profound difficulty making everyday decisions.

lowa Gambling Task

• Damasio developed a clinical test for inappropriate decision tasks referred to as the lowa Gambling Task.
• Individuals turn over cards from four set piles; each displays a loss or gains from points on cards as individual plays
• Two high-risk piles and two low-risk, with the low-risk decks pay better in the long run.
• Control subjects pick high-risk decks to start out
• Those individuals with no neurological conditions choose more low-risk decks over time but the individuals with decision-making deficits will continue on the high-risk decks and lose all of their points.

Somatic Marker Hypothesis

• Emotions inform decision making via bodily reactions.
• Fear would cause a relevant stimulus.
• Galvanic skin responses and increases in heart rate characterize control subjects in IGT, but not Damasio's patients.
• These responses evolve with aiding choices in future.
• Poor can come disrupt somatic markers since not all behavior can influence emotions.
• Scans of Damasio's patients showed there were somatic markers in the vmPFC and damage resulted in damaged decision making skills
• VmPFC also has connections to the amygdala.
• Reconstructions of Phineas Gage’s skull suggest damage to the same area.

Conclusions

• Differences can occur in terms of individual, uncertain or risky outcomes that exist in decision making.
• Research shows that popular belief that emotions should be kept "in check" but emotions may facilitate effective decision making.
• Choice is a topic encompassing animal foraging behaviors through modern economic theory.
• While having strong beliefs in free will, many of our choices are governed by other choices. Individual differences are seen in human and in non-human animals with the task of choice-making.
• Choice plays a critical role in the foraging behaviors of non-human animals.