rolandi@gollum.Columbia.NCR.COM (rolandi) (04/05/88)
In response to Arti Nigel: >The qualifiers mean what you would expect based on standard English meaning: >..... but I think you have a problem with my adjectives! No problem at all. I just needed to determine whether you were using adjectives as opposed to (say) coining some new technical terminology. >>>Most parents do not follow their toddlers around correcting the grammar ... > >>is meant to convey that reinforcement in language acquisition is necessarily >>something that comes from one's parents. >Well, okay, let's say from other people who are capable of responding >differentially (of providing contingent reinf.) to the child's language >production. My nephew is an only child who as an infant was rarely around >people other than his parents for a significant amount of time, so I made >that mistake because I was thinking of him as I wrote the above. OK. Let's make a distinction between reinforcement (a process or act) and reinforcers (stimuli that have reinforcing properties for an organism). Positive reinforcement takes place when a stimulus follows a response and, as a result, the probability of that response is increased. Negative reinforcement takes place when a response is emitted and its emission coincides with the termination of some aversive stimulation. Like positive reinforcement, negative reinforcement increases the probability of the response. While reinforcement may be relatively easy to define, reinforcers, at least outside of a laboratory or other experimental situation, are not always easy to identify. What may act as one person's reinforcer could have a punishing effect on someone else's behavior. Any arbitrary and neutral stimulus can acquire reinforcing properties as a function of paired association with an existent reinforcer. Amazingly enough, even normally aversive stimuli (like electric shock) can take on the characteristics of a positive reinforcer under certain situations. The point is that it is difficult to identify reinforcing stimuli in terms of their inherent properties. Although frequently assumed to be pleasant rewards, or in your case above, a parental correction, a reinforcer need not be appetitive nor even instructive. What makes a stimulus a reinforcer is, rather, its function. Walter Rolandi rolandi@gollum.UUCP NCR Advanced Systems, Columbia, SC University of South Carolina Departments of Psychology and Linguistics
arti@vax1.acs.udel.EDU (Arti Nigam) (04/05/88)
To: rolandi@gollum.UUCP Subject: Re: reinforcement Newsgroups: sci.psychology In-Reply-To: <84@gollum.Columbia.NCR.COM> Organization: University of Delaware Cc: Bcc: In article <84@gollum.Columbia.NCR.COM> you write: > >In response to Arti Nigel: >>The qualifiers mean what you would expect based on standard English meaning: >>..... but I think you have a problem with my adjectives! > >No problem at all. I just needed to determine whether you were using >adjectives as opposed to (say) coining some new technical terminology. > >>>>Most parents do not follow their toddlers around correcting the grammar ... > >>Well, okay, let's say from other people who are capable of responding >>differentially (of providing contingent reinf.) to the child's language >>production. > >OK. Let's make a distinction between reinforcement (a process or act) >and reinforcers (stimuli that have reinforcing properties for an organism). Ah, I see what you mean. Yes, I fully understand the meaning of reinforcement (positive or negative, it increases recurrence of the 'response.') and of punishment and extinction (both of which decrease the recurrence.) And I know that a reinforcer in any given circumstance is defined not by its specific nature but rather by its function/effect on the 'response.' Such a construct (reinforcer as defined in this way) is useful in trying to explain behavior that currently occurs or that has occurred in the past. It is not very useful, I tentatively suggest, in determining what will cause a certain class of behaviors to start occurring. Yes, different things are reinforcers to different people/behaviors under different conditions. But it is mighty interesting, and needs a theory to explain, that virtually every child on earth has an amazing propensity to get reinforced for a certain class of behaviors; very regularly, EVERY child gets reinforced for this class of behaviors (language production); there must be a large body of potential reinforcers relevant to this class, as well; it is unlike many behaviors that some people learn, some don't; so, language is a pretty special thing, and the environment seems well set up to be exceptionally conducive to (provide much and varied reinforcements for) language production. I think that fact is very interesting, and if you want to know WHY rather than just the elements of HOW, you are not asking a fruitless question. You are asking something that strict behaviorists do not claim to be answering (I think). Certainly behaviorists have not succeeded in answering that, and I think they have not attempted it. That is certainly just fine; no one has to know everything; but it is a legitimate scientific question to deal with. Arti Nigam. >Walter Rolandi >rolandi@gollum.UUCP >NCR Advanced Systems, Columbia, SC >University of South Carolina Departments of Psychology and Linguistics
rolandi@gollum.Columbia.NCR.COM (rolandi) (04/06/88)
In response to Arti Nigam: >Such a construct (reinforcer as defined in this way) is useful in trying >to explain behavior that currently occurs or that has occurred in the past. >It is not very useful, I tentatively suggest, in determining what will >cause a certain class of behaviors to start occurring. You are right. The process of shaping however, addresses this issue. >.... But it is mighty interesting, and needs a theory to explain, >that virtually every child on earth has an amazing propensity to get >reinforced for a certain class of behaviors; very regularly, EVERY child >gets reinforced for this class of behaviors (language production); there Why is this so amazing? The same can be said of the class of behaviors associated with locomotion: crawling, walking, and running. >must be a large body of potential reinforcers relevant to this class, as >well; it is unlike many behaviors that some people learn, some don't; >so, language is a pretty special thing, and the environment seems well >set up to be exceptionally conducive to (provide much and varied >reinforcements for) language production. I cannot tell which side of the debate you are on. Is language acquired through reinforcement or not? >I think that fact is very >interesting, and if you want to know WHY rather than just the elements of >HOW, you are not asking a fruitless question. You are asking something >that strict behaviorists do not claim to be answering (I think). Certainly >behaviorists have not succeeded in answering that, and I think they have >not attempted it. That is certainly just fine; no one has to know >everything; but it is a legitimate scientific question to deal with. Was there some point to this last part? What are you saying? Walter Rolandi rolandi@gollum.UUCP NCR Advanced Systems, Columbia, SC University of South Carolina Departments of Psychology and Linguistics