j=e ic,67 'Derr I)r@ Enc2.osed is your which in to be cubmitted f or rubl:Lceion in the j our=l P..,, The article ae edl@ted bas been forml-,,r cler-red for In -11;' 6 ----opet -,crtiture with the proviso tl2at ro Agency n:ffi3-ir-tion v= be revealad. YC=Z 40 OPx-.Mrj, FXIIZPVRCFIEv"&" OF A @=IA= AUTONUIIC RESPOI.'SE Bacl,.rround: ilcr-frer and some other psycholoi; ists have argued that there are two distinct.methods of conditioning or learning (roughly# classical 2 con- ditioning and instrumental learning). Furthermore they have insisted that instrumental learning cannot be reduced to classical conditioniz3g" and that classical conditioning ca=o*** be explained by the laws 0&' instrlmental learning. As evidencej, Yowrer points to the f2act that autonomic responses and skeletal responses are qualitatively different as t7ney are given in nature. "Here.we are ass=ing that behavioral responses are categorica-Ily ditfferent from e=tional responses: the former are 'voluntary' &rA s-i@oject to 4@nf;:.aelice tJaroua@2 reward.and punishment (and not conditionablej, 3trictly speald.-ig)o whereas the latter are involuntary and conditioaa7ole and not a-@ioject to control ame way as are .hrouga reward and puni3Lment,, or at 2Aeast not in +;ae s the overt bebavioral responses." (i4ower., '@.960.) Tiaere are everyday "I&n3-,.ances c@' apparen,,'.,learr4rAa of an autonomic response through reward; SItinner ci-4,;es the child who cried "real tears" because tears 2k-.ad been folladed by itttention and candy in t7ne pastj. and recentl,,f soae experimentrll ev,Adence has sho-vm that ala'@,ono;4ic respor4es can be directly modiried by response cont4&,:39ent reinforcement. however., it is not ,)ossib"@c to rul8e out the effect of skelet"l mediat-'-ng responses in any of these cases. If the autonomic response in questioa is clicired uy c-n une7aservcd skcletal responsc, it is possible that all of the evidence for apparent autonomic learnine merely reflects the parusitic reinforcement of the undcrlyim skeletal response. Some authors (Kandiers 1962) have insisted that even the posalbility of skeletal mediators is sufficient to reject the li@-potnesis of rutoncmic learni2ng through direct reinforcemento' 7'he mediation explanation asserts that the mediator is a skeletal response which precedes and elicits the autonomic response. t..s a rea:ult of rewarding the autonomic response, the skeletal response is parasitically reinforced and learned. Thus it is i=lied that a typical neg&tively accelera2ted learning curve vauld develop for the skeletal response. I4. is also implied that ttle variabili-k,,r in the autonorac response can be 4"uUy exdlained by the changes in the frequency of the underlyizlg skeletal response. In the present experiment,, it i3 proposed to set up an e.%-plicit medi- 2 ating skeletal response which elici@s a drop ir. galv='Lc skin resistance# and then reinforce the e2leited --utcr..=ic response. If the mediation explanation in carre.-t., it should be Possible to discover both the negatively accelerated curve'.of the skeletal response and to expl2ain the variability of the elicited autonomic response. It may also be posr.-I-ble to compere these find.1r@gs with the results o-@O the earlier studies Viere an auton=;-c response was apporen4,.ly dire--tly reinforced. If the learning curves are verv similaro. It vo3uld sugge3t that an unknown skeletal m--diator wai responsible for the chaziees observed in the earlier work. -3- It is possible that results fr= this study will give us some kind of model of operant learning of auton=ic responses as it occurs in stressful situations* Method: The method will be a fairly direct extension of methodology used in earlier studies CoU2eZe a,;.-,e female volunteers wl-11 be used; this will permit a reasonable compari&on with tte.old data. Ten experimental and 10 control subjects will be nm,, they will be matched for frequency of GSR nonepecific defle--tions on the first day o-.P the ekueriment and yoked for schedules of reinfor2cment for the remainder of the experiner,%. 'In the earlier studies it was observed ttat very larf:e respirations occurred intermittently in what wAs othemise a very regular respiration rate. Ttiese large respirations were often followed by a GSR deflec4h-lion. For examplej, on the last da@,, of th2e 196,3 experi=ent,, the eichteen aiibjec4ks emitted 0 to 26 gross irrerularities of respirations during the twenty minute session with an average of It per subject. These respirations often elicited GSH deflections. - 7i@.e rate of GER elicitation varied from 12@p'.@for one of the subjects to 100'@'2) for four of the subjects; the med@Aan was 5u"i'lo tl4a-%-, ey.T,erimerit a CpeLial ef.,ort was racle to a,.-oic@ reiz;!'orcinr, respiration-cli--ited def!L-ctici@3. In ' .Ae it is pro- pored that t.1rese gross irregula--!.ties of7' re-.piration be as a.:elet@al mediators; and tl,,at the4--@- associated Gcift dei'lf--ction,--- be syste=ti- cr-"y reinf'orcee4. 'Xn'L response I)as tht@ &(-.,.,antnee cl' beirg rclativelv subtle and not irawdiately discernible to the subject. it is suggested that the responses be =nitored by the experimenter at this point. It would probably be too cpaplicated to set up an automatic reinforcer for what should be a relatively short,, contained study. otherwioej, methodolov will follow the earlier stadies. All subj2ects will be given'2 days of adaptation, five days of rei nforcement (contingent or yoked) and three dktps of eytinction. We will follow the polic,.%, of having the subjects sit quietly for twenty-five minutes before each day's session of twenty ninuter,. All gii-18 will be screened for medical problems and run in between men2strual periods. The ey-periment shcrild be conducted in a sound-proof constant temperature rocn. Equi=ent-. Sanborn GS'i with Wenger electrodes,, plethvsmorraph (Yenelco Corp.) pno=cm=eter. Data Ana;yais Do you think we can get direct tapinr of t@-ig output of a2ll three variables? In G.4R, all ve need is a frequency co-ant.-&mplitude O'@@' the deflections doe3 not seer.. to be irlro rt..na . Irnat should sir-plit;r the p@-o.o'Le:,-# too. holiever, minute by bcz;a5l re--!.stance coun%"s should be dotained. The fact that basal -esistan--e shifts over the t,.icnty min-a4@.e period may present special difficulties. On thf-3 pletk,,rs=o:,ranh recordp I'd like to take off heart rate (in tweraty second intervals) and amplitude (about, once every twenty seconds)* The respirat.-Lon record suly be the most difficult to automate,, if our experience in recording by hand is any indicator. I have am2le samples of c.11 these, plus a fair2.y specific outline of how they were counted by hand. 2 O.,ice the data are in digital form.. we will need indi-vl&dual and group curves o.-. all...three variables., In the pact 1 havf-- used some rather simole nor-pigra-etric sta-'--Iistics for looking at differences. It ia q@mite clear that the major varilble will be an intra-individaal chanee OLr= 2 adaptation (days 1 and 2) to eytinction (days 9 and 10). It has been suggested to me that there is some vay of lool,.iiig at clusters of all three variables at once as a meas,,--e of learnirc &"). Howevers I am not sure that with this small sample,, non-normal data., etc.# tha"%,o that is feasible. Any suggestions7 see= to me that at this early atage o'L experimen-w.-@6nep the wato- mation of all these things should -ue kepl.; sirxple---ilo leave rocm 'Lor cbanges in procedure and to maintain quite a bit of flex"Lbility in the system. We are handicapped by the fact t2hat our backeround knowledge of ongoinc; basal ratec and i,.idividue-@ d4-fferences i-n GSR nonspecifics., plethy=o,-,raph auplitu:ies., etc. Is still iuite li=ite4. Data Parameters: 1. Basal Rosistance In general, all subjects sl-.m. an i--crease in basu resistance fr= duY one to ton in this kind of stu--y. The coritral.,j tend to show a relatively greater increase than the experimental subjects. alhere are side individual differences in basal resistance. In my 1964 study., I found that it varied from 61.9 K (Wil day 1) to 444.6 K (14HP ftY 5) - Within an individual,, a ma-.cimuru change of aboat 150-200 K could be 2 seen from day one of the experiment to dRy ten. I-led-'Lan readings for eacli day were' used as estimates of each subject's daily level. Basal resistance can shovi rehe-r marre,-'- chanae wit-hin a d@-ily session for ecte i.-idividuals. In the 196); study,, ct leact one subject hz,.,! an increr.se o-140 ov--r. 200 r, f,,-on be,-.inri nr, to entl of the ticiity rir.,Lte aession on dE@r 9 (CJ). Otlier a-Lbjecto tend to rer-.,;.'Ln cr,.iite s-I.-able w'4.thin r- d&yls rur. P-r.1, fr= dr,%,- to dr..v. Ir. eenei-a.!$ sublects with lo-i basal res'istancti pu,4%,, cAx4. a tUC4,li fre- quency of nons:)ecifies,, while subjl-cts i:it:,t high bas&'L resistance proauce fL-.4 nonsnec-Al.-.".-A.21-is. For instance, zi-n tiie l9G4 stud@rj one subjecl,- (@C) had ar. ini-%.ial basal resistance of cn,@ 61 K a.-it! her oi:',p,.t+- o'L no.-.s,,)eci4"-cz was over 6 pe?,- minute. lino-Lher stfuject i-ritai a higii basal resistance 2 (33'4. X) had a nonspecif-i-c ral&-.e o-&' .4 per niml%-e. 2. "ionspeciloic--s In the 1964 study,, experimentals tended to maintain their initial rate of nonspecifics while controls declined. Tl,-,ere are lar,- in fr--quer..-y9 cf noispe-ificc. individua" d!.fference- In the 1964 study,, the love;3t reading on day 1 wai: .4 per mimte,, the high vas over 6 per ninute. Botli experirdentals an@;. controls show some -7- dec3lne in rate over the tan days, so the ran,7e on dmy 10 is -05 per minute to about 5 per minute* I do no'ip have any current. data on amp.Utude of defle--tiorAs. The- amnllest we counted was .4 Y, and the largest was 30 I2C,, but there were ur,quectionab@l.v lamer ones. Data i@rm the 1962 study miggests that size tends to increaue as basal resistaiiee increases fr= day one to day ten; median 9;V3.itude increased at =,cl,. as three times the da@, one arir-lit,.ide (conductance readings). 2 3- Heart Rate In the, 1964 study,, experimentals maintained their initial level of heart rate while controls showed a a-n" but consistent decline. A,gain,p there are fairly large individual differences. In the 1964 study,, the eighteen subjects ranged ftcn 61.8 to 104-7 with a gtandard deviation of 12.8. Under non-streseful repeated trials., the standard deviation within an ir-dividual in about 6 beats per minute. The maxiim:Lm within individual variation that we saw over days was 15 beats I?er minute. 4. Plethysmograph Tllis data is V=ely i-,jil@zea data. On the first duy of the "96)4. stucly 2 'we found that subjects varied fro--, a mini=am =plitude readin4; to twice that readinz. Over tlie ten day sessiotip controls ten,-@,ed to increase L'% a.mplitude vhile experimen-6als rerx,-in re2z.-ively - conslu,.t. Sme slabj ccls doubled their empli-u4ude over the tan dLyt.., v,!iile a4v- leaslt ontt subject sho-,qed a reading which was bc-If a.-. large hi@@ initir-l- amplitude. Cost computer for this study vill be tape and The heaviest mmenses cme estimates at we made a timo Each subject is run 2f6r 3 1/3 ham-8, et know vhether these costs this kind of study- 1-dOn @or @are CC=- amble, but for what tbe7lre;worth Appa=tus batteriero, ctc- Analog2 Recorder tape at ..... . Dieital taps at con- imee Analog to digital, computer t version at al,93.Ysio Lt ccmput 2 er time Digital SubjectS at Total to get up the a computer programer This does not include the cost Of take as long an three months and 2 'We estimated that Might progr=. at about