Welcome to Experimental Biology - PowerPoint PPT Presentation

About This Presentation
Title:

Welcome to Experimental Biology

Description:

Welcome to Experimental Biology Get a syllabus First day policy applies today (take roll) You do need to purchase a book from Target TA s are Sarah Tso, David ... – PowerPoint PPT presentation

Number of Views:54
Avg rating:3.0/5.0
Slides: 21
Provided by: T222
Learn more at: https://www.bio.fsu.edu
Category:

less

Transcript and Presenter's Notes

Title: Welcome to Experimental Biology


1
Welcome to Experimental Biology
  • Get a syllabus
  • First day policy applies today (take roll)
  • You do need to purchase a book from Target
  • TAs are Sarah Tso, David McNutt, Amanda
    Buchanan, Elise Gornish, and Ben Nomann

2
I. Purpose of this Course
  • Teach basic scientific method
  • Create scientific literacy in our biology majors
    to make better citizens
  • Introduce students to research at a Research I
    university (supposed to be for sophomores, but it
    is never too late)

3
II. Scientific Method
  • Some Definitions
  • 1. Science investigation of rational concepts
    that can be tested using observation and
    experimentation
  • -- defined by method of investigation
  • -- limited to the study of physical universe
  • -- must be amenable to experimentation.

4
II. Scientific Method
  • Some Definitions
  • 1. Science
  • 2. Scientific Method formal way of asking and
  • answering questions in science
  • -- common to all fields of science

5
II. Scientific Method
After anyone touches switch, lights come on
Hey, can you turn off the light the same way?
Does the switch on the wall control the lights?
Flip the switch!
Flipping the switch turns on the lights.
6
II. Scientific Method
  • Some Definitions
  • 1. Science
  • 2. Scientific Method
  • 3. Hypothesis any proposed explanation for
    an
  • observed phenomenon, used for further
    exploration
  • -- not a guess, as it is based on prior knowledge
  • -- generally stated as a testable fact

7
II. Scientific Method
So, the scientific method deals with formulating
and testing hypotheses. Things to remember 1.
Science is limited to the physical universe
does not include supernatural forces. Anything
that can't be tested by observation. and
experimentation. is not science, by
definition e.g. A ball falling could be
explained by either Gravity - vs - divine
intervention -- Science can only test one of
these. e.g. "creation science" oxymoron 2.
Doing science uncovering rules that govern
physical universe (vs. facts) when testing
hypotheses. We assume that a given set of
physical conditions will produce a consistent
result
8
II. Scientific Method
Please note that these definitions do NOT suggest
that science and religion are at odds. Instead,
science is based on a formalized methodology that
operates on the physical universe. It simply
cannot be applied to many religious, spiritual
questions, and therefore in general has nothing
to say about the religion. This is related to a
current interesting controversy in evolutionary
biology and philosophy. A variety of atheistic
humanists, principally the biologist Richard
Dawkins and the philosophy Daniel Dennet argue
that belief in Darwins view of evolution
necessarily leads to a belief that there is no
god(s). They suggests that religion is a
cowardly flabbiness of the intellect that
afflicts otherwise rational people.
9
II. Scientific Method
Other prominent scientists and philosophers,
such as Ken Miller and FSUs Michael Ruse, argue
that this view is not only wrong, but is also
dangerous, both at a moral and a legal level.
Ruse goes on to suggest that Dawkins view is
essentially promoting atheism as a religion.
This has important implications for the legal
arguments among creationism, intelligent design,
and evolution, because the current courts are
tossing out intelligent design in schools because
of its religious content, over the supposed pure
scientific view of evolution. -- We wont
resolve this general question in this class. -- I
(along with most major religions in the United
States) feel that evolution has a firm basis in
scientific methodology and should be taught in
schools. Nothing in biology makes sense except
in the light of evolution (T. Dobzhansky).
10
II. Scientific Method
  • Some Definitions
  • How do we test hypotheses in Science?
  • -- by the rejection of alternative hypotheses
  • -- no such thing as proof, only evidence.
  • -- pose reasonable alternative hypotheses
    (often includes a null hypothesis proposing no
    effect).
  • -- design experiment that can discriminate or
    disprove alternate hypotheses.
  • -- experiments should basically disprove all
    hypotheses but one.

11
II. Scientific Method
  • Some Definitions
  • How do we test hypotheses in Science?
  • Lets consider another light example -- a light
    that does not go on when we flip the switch. We
    might suspect that either the lamp or the bulb is
    bad. A good test would be to move that bulb to
    another lamp in the same room and try it again.
    If again the light does not go on, have we proven
    that the bulb, rather than the lamp, is bad?
  • No, there are other possible explanations. For
    example, the power could be out. We have simple
    eliminated the hypothesis that only that lamp is
    bad.
  • Another similar example is the earth and sun --
    who goes around whom?

12
II. Scientific Method
Perhaps a better example is the earth and sun --
who goes around whom? If we assume that the sun
goes around the earth once a day, we can make
some predictions about when the sun will rise and
set that hold up pretty well. We could also even
develop a way to explain seasons, by simply
having the sun take an erratic course around the
earth. So, does this mean that the sun does go
around the earth?
13
II. Scientific Method
  • Some Definitions
  • How do we test hypotheses in Science?
  • -- in short, the problem is that with deductive
    logic you can get the right answer (a successful
    prediction) for the wrong reason.
  • Therefore a more conservative method is favored
  • the rejection of hypotheses that give rise to
  • inaccurate predictions.
  • Nothing is ever proven, only disproved.
  • Science must always be open to new ideas, new
  • alternative hypotheses.

14
II. Scientific Method
  • Some Definitions
  • How do we test hypotheses in Science?
  • Why learn the scientific method?
  • -- it is necessary to make responsible decisions
    about the world around you, especially here in
    Biological Science at FSU.
  • Examples
  • -- Buck-tooth faced baby
  • -- Vioxx, Celebrex, Aleve
  • -- politics, advertising, insurance
  • -- Global Warming

15
II. Scientific Method
Vioxx. Celebrex. Now Aleve. What's a Patient to
Think? By ANAHAD O'CONNOR Published December
28, 2004, New York Times. When Audrey Eisen
flicked her computer on last Monday night and
read the news that the painkiller Aleve had been
linked to heart attacks, she winced in disbelief.
Ms. Eisen, 64, a retired professor who lives in
New York, had just returned from her drugstore
with a package of Aleve. Her pharmacist allowed
her to return it the next morning, no questions
asked. It was the third painkiller in four months
that Ms. Eisen, who has degenerative spine and
disk disease, had quit abruptly because of
studies linking the drugs to heart attacks. She
flushed her Vioxx down the toilet in September,
after it was withdrawn from the market, and
switched to Celebrex. But when problems surfaced
with Celebrex this month, she had to stop that,
too.
16
II. Scientific Method
U.S. approves weight-loss drug for obese dogs
07/01/2007 1407 By Susan Heavey WASHINGTON
(Reuters) U.S. health officials have approved
the first prescription weight-loss drug aimed at
treating Americans increasingly plump pooches.
The drug, Pfizers Slentrol, helps decrease
appetite and fat absorption to help the roughly 5
percent of U.S. dogs that are obese lose weight.
Another 20 percent to 30 percent are overweight
(two-thirds of Americans are also overweight or
obese). Also known as dirlotapide, the
once-daily liquid can also cause various side
effects, including vomiting, loose stools,
diarrhoea and lethargy. Slentrol is not for human
use and will carry warnings to discourage people
from using it, the FDA said.
17
II. Scientific Method
  • Steven Lima, Thomas Valone, and Thomas Caraco.
    1985. Foraging-efficiency -- predation-risk
    trade-off in the grey squirrel. Animal Behaviour
    33155-165.

18
II. Scientific Method
  • Steven Lima, Thomas Valone, and Thomas Caraco.
    1985. Foraging-efficiency -- predation-risk
    trade-off in the grey squirrel. Animal Behaviour
    33155-165.
  • Predict that tendency to carry a food item should
    decrease with distance of food from cover
    (predation risk) and increase with item size
    (food reward). Both risk and reward should
    influence behavior
  • Experiments were conducted in Highland Park in
    Rochester, New York. The reward was
    chocolate-chip cookies, cut to weigh 1, 2, or 3
    g. More natural foods were buried rather than
    eaten and cookies may be a natural food for a
    park squirrel anyway. Food was placed at
    different distances from trees.

19
II. Scientific Method
Steven Lima, Thomas Valone, and Thomas Caraco.
1985.
20
II. Scientific Method
  • Steven Lima, Thomas Valone, and Thomas Caraco.
    1985. Foraging-efficiency -- predation-risk
    trade-off in the grey squirrel. Animal Behaviour
    33155-165.
  • The authors conclude that the results support
    their hypothesis. Simple models that only
    incorporate foraging rate or only exposure to
    predators are insufficient, as both are important.
Write a Comment
User Comments (0)
About PowerShow.com