[prev in list] [next in list] [prev in thread] [next in thread] 

List:       nextgen-online
Subject:    [Nextgen-online] RE: Empirically based teaching methods
From:       rsain () wsu ! edu (Sain, Ryan Christopher)
Date:       2005-11-30 22:03:53
Message-ID: EF6631296121CF4390E090D9B4EFD069014D8DB9 () cru80 ! cahe ! ad ! wsu ! edu
[Download RAW message or body]

Good stuff! 

Paul - I'm not sure we are actually talking about different things. In
fact, I tend to agree with what you have stated. You described things in
observable terms - not hypothetical ones - very good. 

Your example is s good one. Let me explain how it fits with the current
understanding of behavior. Training+Experience. This is exactly correct,
although one could argue that training is experience. Here is where we
need to include the idea of generalization (a concept based on
observation, not philosophy). My car, on a given patch of ice will
behave in a given way. I experience this and modify my behavior to
match. But next year, the snow is a little different, I'm on a different
road, in a different country. Generalization (stimulus generalization)
addresses how one behaves in this instance. Of course, once the old
behavior is used under these new conditions - some things will likely be
modified by the new experience. 

Practical knowledge is a term that describes the ability to do something
- rather than just describe it. So, again, I agree! By the way - were
you in my chemistry lab - I've had a similar experience!~ :-)

- Ryan
-----Original Message-----
From: Paul Scott [mailto:pscott@uwc.ac.za] 
Sent: Wednesday, November 30, 2005 11:39 AM
To: Derek Keats; Sain, Ryan Christopher
Cc: nextgen-online@cvs.uwc.ac.za; tusu@dicts.mak.ac.ug; Wesley Nitsckie;
Beebe, Maria; hmtwaaky@yahoo.com
Subject: RE: Empirically based teaching methods

I *think* what Derek may be getting at here is something like the
following scenario...

Sure, you can train someone to perform complex behaviours, like driving
a car. You can even train them to perform certain behaviours in certain
situations, like steering into a skid. The complexity comes in where the
experience of learning the way that _your_ particular vehicle may or may
not react in certain situations, i.e. I _know_ that my car will skid
because my tires are worn. I have never received training - chained or
otherwise - to cater for different states of wear on my tires that may
exist at any one time, I simply use the knowledge (training +
experience) that I have of my car, the stretch of road, the weather
conditions etc to counter having a fatal accident. I have been trained
in a certain way, but cannot effectively use that training without
experience/knowledge, it is simply not holistic enough to keep me alive.

In an educational sense, training is fine, but practical knowledge also
plays a large part of it. Take chemistry for an example. A lecturer can
train his students to heat alchohol in a water bath, but until the
learner actually does it, and gains the experience of nearly burning
down the lab while monitoring the temperature, there is insufficient
knowledge gained to be useful. I would certainly not employ someone in a
lab that has never lit a burner, but can quote page and paragraph of the
instruction manual.

--Paul

>>> "Sain, Ryan Christopher" <rsain@wsu.edu> 11/30/05 8:31 PM >>>
Derek,

Define something that is a complex behavior! Education is simply
training new responses to new stimuli. We can go into concurrent,
chained and other types of schedules of reinforcement - but I'm not sure
that's what you're getting at. 

Other type of complex responses can be identified as chained. This is a
bit different than shaping. Chained behaviors each have links (a
response) each response serves as a discriminative stimulus (think of
this as a cue) for the next response and as a reinforcer for the
previous. Chaining is likely what happens when you drive - there are
hundreds of behaviors being performed, but it seems like 1, that of
driving. Much the same can be said for a student - going to classes,
etc, are likely chains of multiple behaviors with the entire chain being
reinforced by given stimuli.

Cooperation (generally considered complex) - can be trained in many
species (surprising, this includes humans):-) 

With regard to conscious and unconscious - those are simply hypothetical
constructs (how does one move knowledge [another hypothetical] from
consciousness to unconsciousness [wouldn't that be impossible? how could
we move something to unconsciousness because we would be conscious of
that change - or vice versa]). We cannot observe either - and don't need
to, to describe and modify behavior. Generally speaking, thought about
'doing a behavior' - occurs after the behavior starts (cognitive folk
discovered this - to their dismay). Thoughts can be thought of as
private verbal behavior - under control of various contingencies (have
you ever wanted to say one thing, but said something else?). Helen
Keller had a great point here: Thoughts don't exist without a language.
She described her thought process prior to being able to speak as
"nothing". Cognitive behavior therapy - teaches people to retrain their
thoughts (thinking patterns) while also modifying contingencies to
change their behavior. 

There is nothing special about a complex behavior - it's just many
individual behaviors combined. We can use group contingencies to get
multiple organisms to work on a given task - be that going to the moon
or solving a puzzle. Pigeons were even trained to guide missiles in WW2.


I'm not trying to espouse a philosophy - I'm trying to explain what we
have learned by simply observing behavior and not hypothesizing
constructs to explain it. The issue here is this: if I take a
perspective that a hypothetical construct controls behavior in some way
- then I should be able to modify behavior by modifying that construct -
however since it's hypothetical, how can I modify it? I cannot. Hence
the problem with much of educational psychology - we pose hypotheticals
(mind for example) then the design classroom techniques to modify these
hypotheticals. This is just ludicrous when put to empirical and logical
test. The conclusion is to focus on behavior - something that has been
shown to be affected by consequences. 

If these questions keep coming - I'm just going to create a course on
Learning (and share with everyone!)

Other good reads:

Skinner, BF (1953) Science and Human Behavior.

http://www.behavior.org

http://www.bfskinner.org

Matching Law stuff by Herrnstein (here's one interesting article
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&lis
t_uids=12373433&dopt=Abstract ). 

- Ryan

-----Original Message-----
From: Derek Keats [mailto:dkeats@uwc.ac.za]
Sent: Wednesday, November 30, 2005 7:50 AM
To: Sain, Ryan Christopher
Cc: Paul Scott; nextgen-online; Wesley Nitsckie; Beebe, Maria; Hashim
Twaakyondo; F F Tusubira
Subject: RE: Empirically based teaching methods

Those are all simple behaviours, that are easily moved from consciously
knowing to unconciouslay knowing. How about an example that involves
more complex examples involvving multiple 'stimuli' such as happen in
education (as opposed to training as in the examples you give).

On Wed, 2005-11-30 at 07:01 -0800, Sain, Ryan Christopher wrote:
> Paul,
> 
> You are exactly correct. I will add some more detail:
> 
> 1. The behavior will reoccur in a GIVEN CONTEXT (may or may not 
> generalize). Intermittent reinforcement creates behavior that is much 
> more resistant to extinction and the behavior also occurs at a higher 
> rate. See Schedules of Reinforcement 
> http://employees.csbsju.edu/tcreed/pb/schedules.html . You can read 
> the original here: Ferster, CB & Skinner, BF (1957). Schedules of 
> Reinforcement. Englewood Cliffs, NJ. Prentice Hall ISBN 0137923090 a 
> reprint is available from the B.F. Skinner Foundation 
> http://www.bfskinner.org/books4sale.asp
> 
> 2.  Pretty much sums it up. SHAPING produces 'new' behavior, in a
sense.
> You can shape a pigeon to play table tennis, for example. We are 
> shaped to drive cars, program computers, etc. The basic process is
defined as:
> reinforcing (remember the contingent component here) successive 
> approximations of desired behavior.
> 
> 3. Example. You get reinforced for stopping at a stop light (robot), 
> however, not all stop lights are exactly the same, but you still stop.
> You're getting into the area of STIMULUS CONTROL. A given behavior is 
> reinforced in the context of a given stimulus and not in others, this 
> produces stimulus control over that behavior.
> 
> Once again, the key to understanding Operant conditioning and behavior

> analysis is to understand that we must OBSERVE to know something is 
> happening. I cannot say for instance that reinforcement occurs when an

> organism 'likes' something - the 'liking' is inferred. Further, it is 
> useful to know that we are basically only describing behavior 
> (reinforcement, shaping, punishment, extinction - are all defined 
> empirically - these are not explanatory constructs).
>  
> Good discussion folks! 
> 
> - Ryan
> 
> 
> -----Original Message-----
> From: Paul Scott [mailto:pscott@uwc.ac.za]
> Sent: Tuesday, November 29, 2005 9:41 PM
> To: Sain, Ryan Christopher
> Cc: Derek Keats; nextgen-online; Wesley Nitsckie; Beebe, Maria; Hashim

> Twaakyondo; F F Tusubira
> Subject: RE: Empirically based teaching methods
> 
> On Tue, 2005-11-29 at 14:09 -0800, Sain, Ryan Christopher wrote:
> 
> To summarise:
> 
> 1. Behaviour that is positively reinforced will reoccur; intermittent 
> reinforcement is particularly effective
> 
> 2. Information should be presented in small amounts so that responses 
> can be reinforced ("shaping")
> 
> 3. Reinforcements will generalise across similar stimuli ("stimulus
> generalisation") producing secondary conditioning
> 
> Am I correct? or missing the thread completely? I remember reading a 
> bunch of Skinners articles a few years back, so my memory may be 
> letting me down here.
> 
> --Paul
> 
> 



[prev in list] [next in list] [prev in thread] [next in thread]