How to study behaviour: observation and measurement protocols Flashcards
Why quantify behaviour?
It allows us to examine:
- How behaviour develops
- What causes it
- How it affects an animals ability to survive and reproduce
I.e. adaptive significance - How behaviours evolve
e.g: compare similar behaviours among related spp
Anecdotal descriptions of behaviour is natural history.
Using behavioural observation techniques convert it:
Rigorous, methodical, quantitative description is science
preliminary observations
-> hypotheses
-> ethogram ( a catalogue of different behaviours observed)
hypotheses
e.g. research Q: are males more aggressive than females?
Break it down into testable research objectives (hypotheses)
e.g. do males spend more time on/ initiate/ act more intensely aggressive than females
Hypotheses inform what behavioural data should be collected and how
Ethograms
Inventory of behaviours of a species
Includes:
-description (posture, movements, sounds etc.)
-categorisation
-context
-responses intra/interspecial
-usually for an individual/interactions between a pair
-Rarely complete – usually there is an ‘other’ section
ethograms allow us to determine behaviour
-relative freq.
-durations
-sequences
Ethograms of a single species may change seasonally or geographically
May be gender or ontogenic differences
Describing behaviour
3 main categories:
1) Structure
2) Consequence descriptions
e.g. seals flick sand on themselves resulting in improved thermoregulation – this may not necessarily be the reason they perform this behaviour
3) Relational descriptions
Can be more difficult for some organisms
Unfamiliar species are hard to recognise, some are small, nocturnal, move underwater or use communication methods that are cryptic to us
See: https://www.nathab.com/blog/videos-summer-nights-with-in-sync-fireflies/
Categorising behaviour: a good ethogram has:
–Clear categories
–Exclusive categories (or hierarchical)
–Categories easily recognised in field/lab
–Agreed upon by all observers.
–Behaviour must not alter over time.
–Have detailed descriptions of categories – text, pictures, video etc.
Hierarchies of categories
example:
Mating: attempted copulations and actual copulation
Aggression: F-M, M-F, M-M, F-F
and within these groups are the behaviours threat, chase, fight etc.
Categorising behaviour – coding
Grey seal mothers: 6 mutually exclusive categories:
R: Resting (sleeping & low energy cost non-social behaviours such as comfort moves),
AL: Alert (head up, neck stretched and looking around),
AGNS: Non-sexual aggressive interactions,
IP: interactions with their pup (feeding, flippering, nosing or playing with pup),
L: locomotion (movement resulting in a change of location)
SX :sexual interactions (attempted and successful copulations).
Grey seal pups: 3 mutually exclusive categories:
PR: resting,
PA: active,
PS: suckling (pup’s mouth in contact with mother’s nipple).
Behaviours as events or states
States: behaviours with measurable durations
Events: instantaneous behaviours or the start and end of behavioural states
Cf. Sit vs sitting
Events: provide frequencies of behaviours
States: provide duration of behaviours
Sampling techniques: all occurrence sampling
1: All occurrence sampling:
record all observed instances of behaviour (along with ID & additional info if necessary).
(cf: Ad-libitum sampling)**
Proforma records all occurrences of the behaviour
Records: Time, type of behaviour (FI = fight, CH = chase)
ID of individual performing/ initiating behaviour and recipient ID – using ID no.
outcome (e.g. win, lose or unclear) assigned a number
Limited to: easily observed and rarely occurring behaviours
If there is a chance that you could miss occurrences of behaviours the study data becomes adlib and less useful
Sampling techniques: focal animal sampling
Proforma records the behaviour of a single individual
Records continual data on 1 individual at a time
Records:
- time when the individual begins to perform a behaviour
- Type of behaviour (code)
- The recipient individual
Notes:
The first two columns are the most important
e.g. Male 1 rested (R) 12:13:19 – 12:42:37 when he became alert note: M7 approached him
We know behaviour durations
Very detailed and informative
Limited: difficult to record instantaneous behaviours, hard work for the observer requires focus and observation usually limited to 20-30 minutes before data becomes inaccurate, writing on the proforma requires taking your eyes off your focal individual and taking voice notes requires conversion to written notes after
Sample size is so limited and over brief time periods – v. limited. Diurnal/seasonal variations also impact behaviour.
Another way is to use video and decode after but this can take a long time – months even
Sampling techniques: one-zero sampling
Only records if a behaviour occurs or not in a period of time (e.g. here 1 min intervals)
Score occurrence (1), or non-occurrence (0) of a behaviour during a session.
Allows for longer observations – easier to do
Suitable for more rarely occurring behaviours
Two behaviours can be recorded at once (as in right hand table)
Limitations: does NOT record how many times it occurs
sampling techniques: Instantaneous scan sampling
Similar to one zero but provides more info
Individuals are observed at specific time points e.g. every 5 mins
Limitations: uncertain if behaviours are continuous between time points – gives no idea of duration or frequencies (but gives a time activity budget when recorded over long periods)
More than one individual can be observed at the same time by keeping an order when checking individuals
Allows us to observe synchronies in behaviours
