Methodologies in Behavioural Sciences

Methodologies in Behavioural Sciences and Neuroscience

• Methodologies in behavioural sciences and neuroscience are explored in the study of behaviour.

Simple-System Approach

• Simple behaviours used as indices or markers
• Advantages: Parsimony, ease of replication, and quantification. Reductionistic approach.
• Disadvantages: Missing subtle interactions, and operational/ostensive definitions that lead to meaningless statements.
• Examples: "Rats behaved more," "Animals were more active."

Simple-System Approach: Challenges

• Questions about the approach include:
  • Is it easy to determine if the behaviour is present or not?
  • Is the quantification of occurrences or frequencies sufficient?
  • Is the expression of the behaviour properly measured?
  • Is intensity a factor to consider?
  • Are there other modulations of behaviour to consider?

The 3 Little P's (Fentress, 1988)

• This framework explores the process, pattern, and phenotype in the study of behaviour.
• Process: Organism (straw house) X Environment
• Pattern: Event A (stick house) X Event B (brick house)
• Phenotype: Genes X Experience/Environment

Study of Individual Behavioural Patterns

• Behavioural neuroscience, specifically clinical/medical neurosciences, has unique characteristics compared to other behavioral sciences and neurosciences.
• Direct observations are common, often idiographic, longitudinal, and descriptive.
• Patterns of behavior, whether stereotyped or not, are of particular interest.

Ethograms, Sociograms/Sociomatrices

• Ethogram: A comprehensive inventory and catalog of all behavior patterns within a species. They focus on the form of behavior to study sequences.
• The brain acts as a sequencer of movements/motor actions
• Sociograms: Similar to ethograms, but specifically for social behaviors.
• Ostensive definitions are often used, including detailed descriptions, examples, visual aids, to define behaviours to ensure consistency. This can be challenging for inter-rater reliability.

Studying Behaviour Patterns: Ethograms

• Steps involved in studying behaviour patterns using ethograms include:
  • Choosing the species.
  • Selecting individuals or groups.
  • Choosing the behaviours or signs.
  • Selecting measures.
  • Choosing sampling rules.
  • Choosing recording rules.

Consistency >> Reliability

• Intra-observer consistency/reliability: One observer uses the same method over time.
• Inter-observer consistency/reliability: More than one observer uses the same method; high consistency for high reliability.
• Consistency guarantees replicability.

Define the Behaviour's Boundaries

• Focuses on the segmentation and clustering of behaviours.
• "Bout": a segment/portion/part of behaviour, derived from French word 'un bout.'
• Temporal and spatial boundaries of behaviours need to be precisely defined.
• Clusters of behaviors, actions or events.

Measures of Tendency (FOLD)

• Frequencies: Ratio of occurrences within a fixed amount of time; examples are number of pecks per minute.
• Occurrences: Ratio of occurrences across trials, bouts, or sequences; examples are number of pecks per sequence.
• Latencies: Time between events, states, actions; reaction times. Time for an occurrence.
• Durations: Time a particular event requires. Duration of a single occurrence, total duration, and duration as a proportion or percentage of total time.

Measures of Intensity

• Various operational or ostensive definitions are possible.

Measures of Spatio-Temporal Configuration: Sequences

• Measures of how behaviours are configured over time and space. This includes details like the frequency of elements, and ordering in sequences.
• Includes problems like the Lashley serial order problem.

Kinematics/Dynamics

• Velocity (v)
• Distance (d)
• Time (t)
• Acceleration (a)
• Relevant formulas: d= vt, d = vi + 1/2 at^2, vf = vi + 2at

Sequential Analysis: Methodology and Issues

• Objective: Identifying behavioral patterns (spatio-temporal).
• Randomness: 0% predictability
• Stochastic: > 0% but < 100% predictability
• Deterministic: 100% predictability
• Quantitative tools: Markovian analysis, Information theory analysis, log-linear analysis, time-series analysis.

Corollaries

• How important is the serial order of behaviour? (Lashley, 1951)
• Addressing temporal and spatial structure of sequences is crucial.
• How do preceding events impact current or upcoming events?
• Transitional probabilities (matrices and diagrams)
• Frequencies of trigrams, tetragrams, pentagrams, etc., are relevant.

Stochastic Processes

• Information theory
• Markov chains
• Basic steps:
  • Identifying monograms (types of behaviors).
  • Counting the frequency of these behaviors.
• Examples using behaviour sequence AABAABABAABAAABAB:
• Behaviour A: 11 times
• Behaviour B: 6 times

Step 1: Monograms

• Behaviour sequence: AABAABABAABAAABAB
• Behaviour A: 11
• Behaviour B: 6

Step 2: Digrams and Transition Matrices

• Sequence: AABAABABAABAAABAB
• Matrix:
  • A B
  • A | 5 | 6
  • B | 5 | 0

Step 3: Trigrams

• Sequence: AABAABABAABAAABAB
• Examples:
• AAA: 1
• AAB: 4
• ABA: 5
• BAA: 3

Step N...

• Tetragrams: Example AAAA, AAAB, AABA...
• Pentagrams: Example AAAAA, AAAAB...
• Many analyses stop with digrams because of the understanding that previous events influence current and upcoming events.
• Patterns start to emerge.

Interpretation 1

• Predictive value: patterns? structure?
• Behavioural patterns as indicators of neural patterns.
• Also: FAPs, MAPs, and other pre-programmed motor patterns.
• Universal grammar of movements.
• Contextual analysis: searching for meaning in behaviour (e.g., chickadees communicating).
• Questions about motor cognition (frontal lobes) and sequence learning are raised.

Interpretation 2

• Motor signatures: higher cognitive dimension (planning, optimization of movements, problem solving).
• Cortical regions: higher cognitive dimension.
• Cerebellum: timing, 3D adjustments, fine repetitive behaviours.
• Basal ganglia: sequencing (innate or habit learning).

Sampling: Who or What is Observed and When?

• Ad libitum sampling: "whoever", "whatever", "whenever."
• Behaviour/event sampling: focus on specific behaviours.
• Scan/time sampling: focus on specific time intervals.
• Focal/individual sampling: focus on one individual.

Time Sampling: How is the Behaviour Recorded and/or Scored?

• Continuous recording: all occurrences.
• Instantaneous sampling: behavior occurring at a specific moment? (yes/no)
• Types include point sampling and fixed interval sampling.
• One-zero sampling: is behaviour happening now? (yes/no). Subtypes (convenience, fixed time, random time sampling)

The Microstructure of Behaviour

• This section explores the intricacies of behaviour's detailed structure.

What is Behaviour?

• Behaviour involves changes in time and space.
• Motion or action is central.
• Changes in nervous/endocrine system are proximate causes.
• Terminologies like motivation, emotions, hormones all relate to "motion."

The Linguistic and Musical Metaphors

• Language and music (time and space, like dance) relate to behaviour.
• Some linguistic and musical principles apply to behaviours as stochastic processes.

The Linguistic Metaphor

• Linguist metaphor (from semiotics) is crucial in ethology and neuroethology,
• Syntax (grammar): predictability/patterns/rules
• Semantics: meaning
• Pragmatics: social context
• Prosody: amplitude/intensity
• Question: Is the musical metaphor more appropriate?

The Expression of Behaviour: The Musical Metaphor

• Melody: sequential organization of behaviors; variations/intensity
• Rhythm: temporal organization, patterns, durations
• Harmony: combinatorial organization, coordination/progression
• Trends: sonification of data to musification of data via sound/music.
• Detection of patterns.

Stochastic Processes

• Probabilities in time, dynamic probabilities, transitional probabilities.
• Typical approaches: Markov chains, information theory, time series analysis.
• Using transition matrices for digrams, trigrams, and tetragrams.

The Diversity of Serial Patterns

• Common mistake: assuming behaviour patterns are static and non-dynamic.
• "Prosody" and linguistic metaphor are key considerations.
• Behavioural pattern examples:
  • A always followed by random string, but (a) low intensity is followed 78% of time by BB; (b) A (long duration) is followed 78% of time by BB; (c) A (different tone) is followed 45% of time by C.

Other Factors

• Endogenous factors: metabolism, motivational status, reproductive status, immune system integrity, etc.
• Exogenous factors: stimulus duration/intensity, context/situation, physical context, and social context, etc

Sequences 1

• Simple sequence of mutually exclusive events. Reciprocations (alternating events) are in string 1; perseverations (repeating events) are in string 2.

Sequences 2

• Behavioural events can be discrete but often overlap/temporally spaced: 1. Overlaps; 2. Gaps.

Sequences 3

• Temporal and spatial "invasions" in behavioural events: 1. Overshadowing; 2. Encapsulation.

Sequences 4

• Sizes and shapes of circles represent 1. Intensity, amplitude, magnitude (vertical); 2. Duration (time) (horizontal)

Sequences 5

• Ambiguous events or ambiguous flow of events; 1. Dimensional ambiguity (fuzziness; dashed boundaries); 2. Categorical ambiguity (uncertain identity, e.g., white or grey?); 3. Transitional ambiguity (under-defined boundaries).

Sequences 6

• Very deterministic patterns can exist within sequences that show randomness.

Sequences 7

• Parallelism between two body parts (head & tail), with synchrony (1) or asynchrony (2), demonstrated by a diagram.

Sequences 8

• Reality is complex. Everything is a bit of everything.

Categories of Behaviour

• Observational data (natural behaviours): measures of latency, frequency, duration, intensity and sequencing.
• Learning and cognitive tests: used for testing perceptual and cognitive abilities (perception, memory, decision-making, problem solving, etc).
• States (such as emotions) can be measured using electrophysiological measures (GSR, ECG, etc.) of autonomic nervous system activity.

Clusters (Categories) of Behavioural Assays

• Simple-general—simple reflexes (eye blink, stretch, knee jerk); postures (standing, sitting); locomotion (walking, running); activity levels
• Simple-specific—species-specific (innate/instinctive behaviours), simple/complex action sequences
• Acquired behaviours—reflex/startle/orienting responses, habituation, sensitization, and conditioning (classical/Pavlovian/respondent/type I, instrumental/Skinnerian/operant/type II)
• Developmental—developmental milestones, confounding variables (handling time of day, nutrition, litter size), post-natal effects
• Affective/conative—emotional reactions (freezing, defecation), escape/defense behaviours (biting, hissing), affiliative behaviours (huddling)
• Socio-affective—interactive behaviours (dyads, triads), parental behaviours (maternal, paternal)
• Cognitive—attention, memory, problem solving
• Psychopathological/neurological—preclinical models of disorders

Cluster 1: Simple-General

• Simple reflexes: Stretch, Knee Jerk, Eye blink
• Postures: Standing, Rearing, Lying, Balancing, Sitting, Urination posture.
• Locomotion: Walking, Creeping, Running, Crawling, Swimming, Stalking, Flying, Hopping, etc
• Activity levels

Locomotor Assays

• Direct observation
• Photocell-based systems and other automated systems
• Motor Coordination and Balance.
  • Balance with rotarod
  • Beam walking
  • Footprint analysis

Cluster 2: Simple-Specific

• Species-specific action patterns (innate/instinctive behaviour)
• Simple/complex action sequences
• Normal/abnormal stereotyped behaviours (e.g., perseverations)

Cluster 3: Acquired Behaviours

• Reflexes/startle/orienting responses
• Simple, non-associative learning (habituation, sensitization)
• Conditioning (classical/Pavlovian, instrumental/Skinnerian/operant)
• Includes behaviours like taste aversion, maze learning, etc.

Complex Clusters

• Cluster 4: Developmental cluster
• Cluster 5: Affective/conative cluster (emotion/motivation).
• Cluster 6: Socio-affective cluster (interactive behaviour; incl. parental behaviours: maternal, paternal).
• Cluster 7: Cognitive cluster (attention, memory, problem-solving)
• Cluster 8: Preclinical models of neurological and psychiatric disorders (e.g., Parkinson's, lesions, toxins, 6-OHDA, MPTP, global/focal cerebral ischemia, etc.)

Developmental Clusters

• Developmental milestones
• Confounding variables (e.g., handling time, nutrition, litter size)
• Post-natal and maternal effects

Developmental Milestones in Rodent

• Physical landmarks (eye opening, incisor eruption, fur development).
• Locomotor behaviours (head elevation, forelimb & shoulder elevation, pivoting, crawling, walking, swimming).
• Reflexes (see next slide)

Milestones: Reflexes

• Surface righting
• Air righting
• Negative geotaxis
• Cliff avoidance
• Visual placing
• Limb placing
• Vibrissa placing
• Auditory startle
• Tactile startle
• Crossed extensor reflex
• Rooting reflex
• Grasp reflex
• Bar holding
• Level screen test
• Vertical screen test

Milestones: Assessment

• Species-specific assessments based on documented values.
• Average and range of response age in days for each category and milestone, including species and strains.

Assays for Motivation

• Eating/drinking
• Food intake measures
• Food preference tests
• Reproductive/sexual behavior (mounts, intromissions, ejaculations, lordosis)
• Hormone replacement
• Castration/ovariectomy experiments

Assays for Parental and Social Behaviours 1 & 2

• Observational/descriptive: nest construction, retrieval, licking of pups, actively hovering, motionless positioning—visual examples provided in the slides.
• Experimental: hormonal induction of maternal/paternal behavior, nest construction, maternal/paternal preferences, conditioned place preference, T-maze and Y-maze tests, "Pupomat" Carousel Preference test

Stress, Anxiety, Depression

• Stressors: restraint, electric foot shock, swim, social isolation, resident/intruder, maternal separation, sleep deprivation.
• Measures: Hormones, behaviours, etc.
• Construct validity
• Discriminant validity
• Other related constructs: anguish, arousal, fear, phobia, panic, trauma/PTSD, grief

Rodent Models of Depression

• Using stress to induce depression-like states
• Behavioural despair (immobility), learned-helplessness paradigms.
• Derived from forced swim tests.
• Include dogs, with Seligman's procedure (inescapable shocks in shuttle box), and tail suspension tests in mice.

Problems

• No animal models mimic all symptoms of depression.
• All current models are reactive models.
• Focus typically on psychomotor activity reduction, anhedonia, neuroendocrine responses, cognitive changes, eating/sleeping dysfunctions.

Reactive Depression

• Triggered by stress (social/physical), involving the HPA axis
• Depression involves the HPA axis
• Symptoms in depressed individuals are associated with an enlarged pituitary and adrenal glands.
• Higher levels of glucocorticoids (cortisol and corticosterone) in those with depression, potentially as a result in ACTH changes.

Stress Affects

• Endocrine system: adrenaline and noradrenaline (SAM axis), norepinephrine, acetylcholine, GABA.
• Depressed individuals have higher CRF levels in cerebrospinal fluid (CSF) and more CRF-producing cells in the hypothalamus (compared to non-depressed individuals).
• HPA axis activity is linked to glucocorticoids (cortisol).
• ECT and antidepressants reduce CRF levels

Dexamethasone (DEX)

• Synthetic glucocorticoid, used for a challenge to induce a strong down-regulation of CRF and ACTH.
• This down-regulation typically does not occur in depressed individuals.

Cortisol and Depression

• Graphs illustrate data and differences in cortisol levels between depressed/normal individuals.

Animal Tests of Anxiety

• Tests of anxiogenic/anxiolytic drugs, hormones, and situations.
• Measures include social interaction, light/dark exploration, elevated plus-maze, open field test, defensive burying, thirsty rat conflict.

Cognitive Testing in Animals

• Increased interest in cognitive impairments and neurodegenerative conditions, neurotoxins, and endocrine disruptors.
• Ample use and testing of animal learning and cognition.

Cognitive Processes

• Sensation and perception (psychophysics)
• Attention, working memory, latent inhibition, sign-tracking
• Learning, memory, acquisition, retention (storage), retrieval
• Types of consciousness: implicit/unconscious/incidental, explicit/conscious
• Problem-solving, decision-making

Tests: A Few Examples

• T and Y mazes (typically for invertebrates and lower vertebrates)
• Radial-arm mazes (4, 6, 8, 12 arms)
• Morris water maze
• Detection learning (yes/no, go/no-go to presence/absence of S)
• Discrimination learning
• Identification learning
• Rule learning/learning sets

Preclinical Models of Neurological and Psychiatric Disorders: Sample

• Parkinson's disease models (mutant strains, lesions, toxins, 6-OHDA, MPTP).
• Global/focal cerebral ischemia, photochemical cortical lesions, fluid percussion model (traumatic brain injury), neuropathic pain.
• Dural inflammation (migraine), painful diabetic neuropathy (STZ rat model), nociception, experimental autoimmune encephalomyelitis (EAE), amyotrophic lateral sclerosis (ALS) models, chemoconvulsant models of chronic spontaneous seizures.