A working definition of "behavior" is a necessary aspect of its study. So, what is behavior?

Some specific examples that will come up during the semester

Stimulus-response paradigms

Movements (grouping, spreading apart, migrations)

Social behavior

Territoriality (anti-social behavior?)

Mating

Cooperation

Parental Care

Some perspectives on the study of behavior

Within the realm of general biology

From molecules to ecosystems, where does behavior fit in?

 

Within the realm of behavioral biology

Two ways to think about behavior: How ("Proximate" causes) and Why ("Ultimate" causes)?

"How" questions examine the mechanisms and processes that produce behavior

Involves the interaction of neural networks, muscles, sensory organs, etc….

The realm of neurobiology/physiology; ethology/psychology, animal communication, the genetics of behavior

"Why" questions examine the adaptiveness of behavior ("the survival value of behavior")

Involves the way behaviors influence survival, resource acquisition, and reproduction.

The realm of behavioral ecology, sociobiology, and behavioral evolution

Example: the migratory behavior of birds:

How: Changes in daylength trigger hormone shifts that induce birds to fly north

Why: Seasonal peaks of food availability make a migration more profitable than remaining in the south

 

We will explore both, with an emphasis on the latter.

Irrespective of the questions you ask about behavior, it is important to realize that behaviors are shaped by natural selection:

A quick review of evolutionary theory: the importance of THE BIG THREE: Food, Sex, and Death

Heritable variation in the ability to produce offspring causes selection for certain traits that are passed on to offspring. These can be behavioral, as well as morphological.

 

A quick review of the approaches to studying behavior (and biology)

 

Generating and testing hypotheses

  I. The basic goal and methodology for studying animal behavior

The goal: to explain how or why a particular behavior is expressed

Modes of attacking this problem: an example of behavior and thinking about its origins: Sage Grouse displays

1) Not so good: Ad Hoc (special purpose) reasoning: Developing a single explanation for a phenominon that may appear plausible, but is untestable

Problem: Tests using this approach are the original data that suggested the hypothesis... there is no independent verification.

2) Better: Predicting causality: Testing many predictions from a single hypothesis (A leads to B, which leads to C, etc....)

Problem: Ignores the possiblitiy of alternative explanations that may be the true cause.

3) Best: Strong Inference: Discrimination among alternative hypotheses

Using strong inference

1) Generate a set of hypotheses based on observations

2) Strive to make hypotheses exhaustive and exclusive

3) Develop tests to reject one or more of the hypotheses

4) Repeat the sequence, using the results of previous tests to generate new hypotheses

 

Things that limit the effectiveness of using a strong inference approach

1) Insufficient observations may limit hypotheses generation.

2) Bad hypotheses may arise from poor understanding of evolutionary theory

3) Setting up a list of hypotheses that is both exclusive and exhaustive is difficult

4) Rejecting hypotheses can be difficult with lots of variance or small sample sizes (e.g., insufficient statistical clarity)

 

Traditional ways to study animal behavior.

Come up with a single plausible explanation... the old school approach.

1) Benefits: sufficiently diverse data can suggest non-intuitive hypotheses

2) Problems: as above, this approach only identifies a single hypothesis; it must then be proved or disproved independently with additional data...

3) Primarly a violation of the strong inference protocol: hypothesis list is not exhaustive

 

The comparative/observational method... looking for correlations among variables.

1) Benefits: can reject hypotheses and reduce the list of possible explanations

2) Problems: this approach only proves association, not causation

3) Primarly violation of protocol: hypothesis list is not exhaustive and sufficient testing of hypotheses often lacking.

 

Model building... predicting relationships among variables based on a basic set of assumptions.

1) Benefits: this approach may suggest non-intuitive possibilities

2) Problems: the assumptions of the model and number of variables are often unrealisitc

3) Primarly violation of protocol: hypothesis list is often neither exclusive nor exhaustive.

 

Field Manipulations/Experiments

1) Benefits: this approach can prove causal relationships among variables

2) Problems: field experiments are often difficult and negative results are difficult to interpret

3) Primarly violation of protocol: knowing whether your experiment and controls are actually doing what you think.