From db30+ at andrew.cmu.edu Tue Feb 2 12:00:08 1999 From: db30+ at andrew.cmu.edu (Daniel J Bothell) Date: Tue, 2 Feb 1999 12:00:08 -0500 (EST) Subject: ACT-R Environment for ACL 5.0 Message-ID: The ACT-R Environment for Allegro Common Lisp 5.0 for Windows is now available from the ACT-R Environment link on our web site: http://act.psy.cmu.edu/ or directly at: http://act.psy.cmu.edu/ACT/ftp/education/index.html . To use it you must have a version of ACL 5.0 other than the free one. The stand-alone application is not yet ready. If you have any questions or problems with it, please send them to me (db30 at andrew.cmu.edu). Thank you, Dan From db30+ at andrew.cmu.edu Tue Feb 2 12:26:24 1999 From: db30+ at andrew.cmu.edu (Daniel J Bothell) Date: Tue, 2 Feb 1999 12:26:24 -0500 (EST) Subject: New Tutorial Projects Message-ID: Due to changes in the ACT-R Scripting Extensions (which were made for the benefit of the Windows ACL version of the Environment) there is a new set of Projects to accompany the online ACT-R Tutorial. The Tutorial has been edited to reflect the changes, and the Environments have been updated. Some of the new project files will not work with the old Environments and vise versa. Therefore, if you are using the Tutorial it is recommended that you obtain the latest version of the Projects and the Environment. The Projects and Environments are available from the main ACT-R page under the ACT-R Environment link at: http://act.psy.cmu.edu/ or directly at: http://act.psy.cmu.edu/ACT/ftp/education/index.html . The Tutorial is available at: http://bk1.psy.cmu.edu/inter/ACT-R-tutorial.html . If you have any questions about the changes, problems obtaining the new versions, or trouble with the new software please let me know. Thank you, Dan (db30 at andrew.cmu.edu) From cl+ at andrew.cmu.edu Tue Feb 2 16:57:38 1999 From: cl+ at andrew.cmu.edu (Christian J Lebiere) Date: Tue, 2 Feb 1999 16:57:38 -0500 (EST) Subject: Scripting Extension 2.1 Message-ID: As Dan mentioned, we had to implement a number of changes to the scripting extension used in the web tutorial projects to alleviate naming conflicts between scripting commands and Allegro CL 5.0 functions. The release notes detailing these changes are included with the scripting extension and are appended below. A new stand-alone version of the scripting extension is available on the ACT-R home page (act.psy.cmu.edu) by following the Scripting Extension link. Feel free to contact me if you have any questions. Hopefully concluding this flurry of announcements, Christian ---------------------------------------------------------------------- Release Notes for Scripting Extension Version 2.1 ================================================= The Scripting Extension Version 2.1 includes the following name changes of the user-level functions, because of incompatibility with Allegro CL 5.0: invert -> transpose array -> matrix sequence -> gather The following internal function has also been renamed to avoid conflict with the optimizer: square -> square-data The manual and examples folders have also been updated. From gaj at psychology.nottingham.ac.uk Thu Feb 4 09:14:19 1999 From: gaj at psychology.nottingham.ac.uk (Gary Jones) Date: Thu, 4 Feb 1999 14:14:19 +0000 Subject: Parameter settings Message-ID: Hello all, I'm examining the effect that different mechanisms of development have upon a model of adult behaviour on a developmental task. Some mechanisms are particularly suited to ACT-R (e.g. strategy choice can be operationalised using expected gain noise). So for example, as you increase EGN, the behaviour of the model is degraded (fitting the reaction time data of seven year old's on the task). Parameters such as EGN and retrieval threshold are open-ended though - are there any specific ranges that are normally adhered to? (Catastrophic changes in behaviour are not seen in the model until EGN=5 or 6; I expect the "standard" range for modelling adults would be 0-1). Gary Jones Psychology Department University of Nottingham Nottingham NG7 2RD England From ja+ at CMU.EDU Thu Feb 4 11:10:30 1999 From: ja+ at CMU.EDU (John Anderson) Date: Thu, 4 Feb 1999 11:10:30 -0500 (EST) Subject: Parameter settings Message-ID: Unfortunately, I don't think this is one of the cases where we have much guidance for parameter setting. The meaning of EGN (or EGS) is relative to the goal value which multiplies the probability (PG) and to the values one chooses to set C. Basically it is a way of scaling the differences in production utilities. So if, in one model, a production A has utility 19 and another 18 and EGS = 1, one will get the same choice behavior as in another model where a production has utility 30 and another 27 and EGS = 3. The only qualification to this idea is that if utilities get low enough that they might go below 0 one has to worry about scaling distance from 0. If all productions have utilities below 0 the goal pops with failure. That being said, I would guess that for a normal application with G = 20 and P = 1, and the differences being those that do occur in C, Gary is right in the 0-1 range he intuits. I would intuit that range for EGS not EGN if I wanted to state my intuitions precisely. Remember that EGN is equal to EGS squared times 3.29. EGN is the variance and EGS is a parameter of the logistic distribution which is proportional to standard deviation. Excerpts from mail: 4-Feb-99 Parameter settings by Gary Jones at psychology.no > I'm examining the effect that different mechanisms of development have upon > a model of adult behaviour on a developmental task. Some mechanisms are > particularly suited to ACT-R (e.g. strategy choice can be operationalised > using expected gain noise). So for example, as you increase EGN, the > behaviour of the model is degraded (fitting the reaction time data of seven > year old's on the task). Parameters such as EGN and retrieval threshold are > open-ended though - are there any specific ranges that are normally adhered > to? (Catastrophic changes in behaviour are not seen in the model until > EGN=5 or 6; I expect the "standard" range for modelling adults would be > 0-1). From cl+ at andrew.cmu.edu Thu Feb 4 11:13:02 1999 From: cl+ at andrew.cmu.edu (Christian J Lebiere) Date: Thu, 4 Feb 1999 11:13:02 -0500 (EST) Subject: Parameter settings Message-ID: Excerpts from mail: 4-Feb-99 Parameter settings by Gary Jones at psychology.no > I'm examining the effect that different mechanisms of development have upon > a model of adult behaviour on a developmental task. Some mechanisms are > particularly suited to ACT-R (e.g. strategy choice can be operationalised > using expected gain noise). So for example, as you increase EGN, the > behaviour of the model is degraded (fitting the reaction time data of seven > year old's on the task). Parameters such as EGN and retrieval threshold are > open-ended though - are there any specific ranges that are normally adhered > to? (Catastrophic changes in behaviour are not seen in the model until > EGN=5 or 6; I expect the "standard" range for modelling adults would be > 0-1). I argue in my thesis (now available in .ps and .pdf format at http://reports-archive.adm.cs.cmu.edu/anon/1998/abstracts/98-186.html) that catastrophic behavior happens when the activation noise value t (see terminology note at the end to distinguish between t, s, and variance) goes over 1. The value that best fits the cognitive arithmetic data (and can be seen as optimal - see Chapter 5 of the thesis) is s=0.25. That value has worked well for other simulations (as has the value s=0.5). That seems to be the general range for activation noise. The game playing model presented by Dieter Wallach at last year's workshop also used an expected gain noise value s=0.5. However, in general the expected gain scale is going to be proportional to the value of G, and thus one wouldn't a priori expect a narrow range of egs values to fit for all possible values of G. The Atomic Components of Thought book (and the ACT-R web site) presents a number of models that use expected gain noise (check the index for Noise in production evaluation), and but for one exception they all have a t value below 1. However, they also tend to have very small G values in the range of 1 to 3. Regarding the retrieval threshold, Chapter 7 of the ACT book on List Memory uses a wide range of values (from less than 0 to more than 3), but determines a close speed-accuracy-like relationship between the retrieval threshold and the latency factor. Thus even when parameters vary widely they seem to do so with some sort of regularity, suggesting possible further restrictions. Terminology note: :egn (and :an) sets the variance of the noise. While those keywords still work, the new standard is to set :egs (and :ans), the s value of the noise, which is related to the variance by s=sqrt(3*variance)/pi. The temperature t used in the boltzmann equation to determine the probabilities of choice is directly related to s by t=sqrt(2)*s. From gray at gmu.edu Fri Feb 5 12:20:44 1999 From: gray at gmu.edu (Wayne Gray) Date: Fri, 5 Feb 1999 12:20:44 -0500 Subject: ACT-R Workshop Message-ID: --============_-1293886580==_ma============ Content-Type: text/plain; charset="us-ascii" SIXTH ANNUAL ACT-R WORKSHOP =================================================== George Mason University - August 6 to 9 1999 =================================================== ACT-R is a cognitive theory and simulation system for developing cognitive models for tasks that vary from simple reaction time to air traffic control. Each year a workshop is held to enable current users to exchange results and ideas. The sixth Annual ACT-R Workshop will be held at George Mason University, outside of Washington, DC from August 6 to 9 1999. Current users, potential users, and knowledgeable observers of cognitive science are invited to attend. The workshop will take place from the evening of Friday, August 6 through the afternoon of Monday August 9. The pre-workshop will begin Friday morning with a one-day tutorial introduction to ACT-R. The tutorial is suitable for users of other modeling systems as well as for knowledgeable observers of cognitive science. (There is an additional $75 fee for the tutorial.) A reception and poster session will be held Friday evening. Saturday, Sunday, and Monday mornings will be devoted to a series of research presentations, each lasting from 15 to 30 minutes. Afternoons will feature discussion sessions and instructional tutorials. Evenings will be occupied by demonstration sessions during which participants can gain a more detailed knowledge of the models presented and engage in unstructured discussions. Admission to the workshop is open to all. (The registration fee will be announced at a later time. Reduced rates will be available to students.) KEY DATES: All papers, posters, demonstrations, and tutorial attendees: a notice of "intent" to participate is due ASAP but at least before June 15th. Posters and Demonstrations: Abstract due July 1st. One-page, camera-ready copy due August 1st. Papers: Two-page abstract due June 15th. Camera-ready copy is due August 1st. Registration: Dates for early and late registration will be announced at a later time. (http://hfac.gmu.edu/~actr99) Attendees are encouraged to submit by email a notice of "intent to participate" as soon as possible but prior to June 15th. The notice should be short (less than one email page or < 250 words) and should contain the following information. Category (spoken paper, poster, demonstration, tutorial, or observer). Title, description, and other details are optional as long as the total length is less than 250 words. The purpose of the "intent to participate" is to aid the organizers in planning. It will not be regarded as a commitment to attend. Spoken Papers: Spoken presentations can be up to 30 min in length. Researchers are invited to submit a 1-2 page extended abstract (500-1000 words) by June 15th. Preference will be given to running ACT-R models, ACT-R as a modeling language, modeling issues or techniques, models with empirical data, or empirical data pertaining to some aspect of the ACT-R theory. These papers will be reviewed and published (as is or in longer form) in the workshop proceedings. Selections will be announced by July 15th. Up to 8 pages will be published in the proceedings. The 8 pages may include an abstract plus slides or a complete paper. Camera-ready copy must be submitted by August 1st if it is to be included in the workshop proceedings. (The format for the extended abstracts and papers will follow that used for the Cognitive Science Conference. Templates will be available from http://hfac.gmu.edu/~actr99 at a later time.) Posters: All attendees may give a poster presentation. Poster abstracts will not be reviewed. A notice of "intent to participate" should be sent as soon as possible but prior to June 15th. A one page abstract will be published in the proceedings if submitted prior to August 1st. (The format for the abstracts will follow that used for the Cognitive Science Conference. Templates will be available from http://hfac.gmu.edu/~actr99 at a later time.) Demonstrations are solicited for running models, power tools for modelers, or other types of executable systems of interests to the modeling community. A notice of "intent to participate" should be sent as soon as possible but prior to June 15th. A one page abstract will be published in the proceedings if submitted prior to August 1st. (The format for the abstracts will follow that used for the Cognitive Science Conference. Templates will be available from http://hfac.gmu.edu/~actr99 at a later time.) Suggestions for the topics of the discussion sessions or for tutorials on some aspect of ACT-R are welcomed. Topics do not have to be limited to ACT-R, but may include general issues dealing with some aspect of computational cognitive modeling or unified theories of cognition. On campus housing and reduced rates at local hotels will be available. Details concerning the registration for the workshop and for housing will be announced at a later date. Additional information (detailed schedule, etc.) will appear on the Human Factors and Applied Cognitive Web site (http://hfac.gmu.edu/~actr99) when available or can be requested at: Addresses for submissions and for more information: 1999 ACT-R Workshop George Mason University Psychology Department Human Factors & Applied Cognitive Program MSN 3F5 Attn: Jane Blechman Fairfax, VA 22030 Fax: +1 (703) 993-1330 Tel: +1 (703) 993-2104 Email: jblechma at gmu.edu Web site (http://hfac.gmu.edu/~actr99) _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ Wayne D. Gray HUMAN FACTORS & APPLIED COGNITIVE PROGRAM SNAIL-MAIL ADDRESS (FedX et al) VOICE: +1 (703) 993-1357 George Mason University FAX: +1 (703) 993-1330 ARCH Lab/HFAC Program ********************* MSN 3f5 * Work is infinite, * Fairfax, VA 22030-4444 * time is finite, * http://hfac.gmu.edu * plan accordingly. * _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ --============_-1293886580==_ma============ Content-Type: text/enriched; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Courier_NewFFFF,0000,0000SIXTH ANNUAL ACT-R WORKSHOP Courier_New= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D 0000,7777,0000George Mason University - August 6 to 9 1999 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D ACT-R is a cognitive theory and simulation system for developing cognitive models for tasks that vary from simple reaction time to air traffic control. Each year a workshop is held to enable current users to exchange results and ideas. The sixth Annual ACT-R Workshop will be held at George Mason University, outside of Washington, DC from August 6 to 9 1999. Current users, potential users, and knowledgeable observers of cognitive science are invited to attend. The workshop will take place from the evening of Friday, August 6 through the afternoon of Monday August 9. The pre-workshop will begin Friday morning with a one-day tutorial introduction to ACT-R. The tutorial is suitable for users of other modeling systems as well as for knowledgeable observers of cognitive science. (There is an additional $75 fee for the tutorial.) A reception and poster session will be held Friday evening. Saturday, Sunday, and Monday mornings will be devoted to a series of research presentations, each lasting from 15 to 30 minutes. Afternoons will feature discussion sessions and instructional tutorials. Evenings will be occupied by demonstration sessions during which participants can gain a more detailed knowledge of the models presented and engage in unstructured discussions. Admission to the workshop is open to all. (The registration fee will be announced at a later time. Reduced rates will be available to students.) KEY DATES: All papers, posters, demonstrations, and tutorial attendees: a notice of "intent" to participate is due ASAP but at least before June 15th. Posters and Demonstrations: Abstract due July 1st. One-page, camera-ready copy due August 1st. Papers: Two-page abstract due June 15th. Camera-ready copy is due August 1st. Registration: Dates for early and late registration will be announced at a later time. (http://hfac.gmu.edu/~actr99) Attendees are encouraged to submit by email a notice of "intent to participate" as soon as possible but prior to June 15th. The notice should be short (less than one email page or << 250 words) and should contain the following information. Category (spoken paper, poster, demonstration, tutorial, or observer). Title, description, and other details are optional as long as the total length is less than 250 words. The purpose of the "intent to participate" is to aid the organizers in planning. It will not be regarded as a commitment to attend. Spoken Papers: Spoken presentations can be up to 30 min in length. Researchers are invited to submit a 1-2 page extended abstract (500-1000 words) by June 15th. Preference will be given to running ACT-R models, ACT-R as a modeling language, modeling issues or techniques, models with empirical data, or empirical data pertaining to some aspect of the ACT-R theory. These papers will be reviewed and published (as is or in longer form) in the workshop proceedings. Selections will be announced by July 15th. Up to 8 pages will be published in the proceedings. The 8 pages may include an abstract plus slides or a complete paper. Camera-ready copy must be submitted by August 1st if it is to be included in the workshop proceedings. (The format for the extended abstracts and papers will follow that used for the Cognitive Science Conference. Templates will be available from http://hfac.gmu.edu/~actr99 at a later time.) Posters: All attendees may give a poster presentation. Poster abstracts will not be reviewed. A notice of "intent to participate" should be sent as soon as possible but prior to June 15th. A one page abstract will be published in the proceedings if submitted prior to August 1st. (The format for the abstracts will follow that used for the Cognitive Science Conference. Templates will be available from http://hfac.gmu.edu/~actr99 at a later time.) Demonstrations are solicited for running models, power tools for modelers, or other types of executable systems of interests to the modeling community. A notice of "intent to participate" should be sent as soon as possible but prior to June 15th. A one page abstract will be published in the proceedings if submitted prior to August 1st. (The format for the abstracts will follow that used for the Cognitive Science Conference. Templates will be available from http://hfac.gmu.edu/~actr99 at a later time.) Suggestions for the topics of the discussion sessions or for tutorials on some aspect of ACT-R are welcomed. Topics do not have to be limited to ACT-R, but may include general issues dealing with some aspect of computational cognitive modeling or unified theories of cognition. On campus housing and reduced rates at local hotels will be available. Details concerning the registration for the workshop and for housing will be announced at a later date. Additional information (detailed schedule, etc.) will appear on the Human Factors and Applied Cognitive Web site (http://hfac.gmu.edu/~actr99) when available or can be requested at: Addresses for submissions and for more information: 1999 ACT-R Workshop George Mason University Psychology Department Human Factors & Applied Cognitive Program MSN 3F5 Attn: Jane Blechman =46airfax, VA 22030 =46ax: +1 (703) 993-1330 Tel: +1 (703) 993-2104 Email: jblechma at gmu.edu Web site (http://hfac.gmu.edu/~actr99) _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ Wayne D. Gray HUMAN FACTORS & APPLIED COGNITIVE PROGRAM SNAIL-MAIL ADDRESS (FedX et al) VOICE: +1 (703) 993-1357 George Mason University FAX: +1 (703) 993-1330 ARCH Lab/HFAC Program ********************* MSN 3f5 * Work is infinite, *=20 =46airfax, VA 22030-4444 * time is finite, * http://hfac.gmu.edu * plan accordingly. * _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ --============_-1293886580==_ma============-- From gt4747c at prism.gatech.edu Fri Feb 5 17:27:33 1999 From: gt4747c at prism.gatech.edu (Scott Peterson) Date: Fri, 05 Feb 1999 17:27:33 -0500 Subject: contributed model Message-ID: I would like to announce that a new model has been added to the ACT-R Contributed Models page (http://act.psy.cmu.edu/ACT/ftp/contributions). This is an ACT-R 3.0 model of the visual enumeration task that demonstrates emergence of the subitizing phenomenon. The model is fully described in a paper I have written with Tony Simon titled "Computational Evidence for the Subitizing Phenomenon as an Emergent Property of the Human Cognitive Architecture," which is in the queue for publication in Cognitive Science. The abstract for our paper is appended below. An earlier version of the model (Peterson, Simon, & Morton, 1997) was published in the Proceedings of the Cognitive Science Society. Your comments are welcomed and appreciated; we hope to make many improvements as we upgrade the model for ACT-R version 4. Abstract -------- A computational modeling approach was used to test one possible explanation for the limited capacity of the subitizing phenomenon. Most existing models of this phenomenon associate the subitizing span with an assumed structural limitation of the human information processing system. In contrast, we show how this limit might emerge as the combinatorics of the space of enumeration problems interacts with the human cognitive architecture in the context of an enumeration task. Subitizing-like behavior was generated in two different models of enumeration, one based on the ACT-R cognitive architecture and the other based on the principles of parallel distributed processing (PDP). Our results provide good qualitative fits to results obtained in a variety of empirical studies. ******************************************* Scott Peterson School of Psychology Georgia Institute of Technology Atlanta, GA 30332-0170 phone: (404) 894-7314 e-mail: gt4747c at prism.gatech.edu ******************************************* From cl+ at andrew.cmu.edu Wed Feb 10 13:14:52 1999 From: cl+ at andrew.cmu.edu (Christian J Lebiere) Date: Wed, 10 Feb 1999 13:14:52 -0500 (EST) Subject: Post doctoral position Message-ID: There is a post doctoral position open to work on the ACT-R theory of skill acquisition, particularly with respect to one of the following topics -- dynamic problem solving, eye movements over problem arrays, or fMRI studies of complex tasks. Contact John Anderson at ja+ at cmu.edu. From tkelley at hel4.arl.mil Wed Feb 10 16:49:30 1999 From: tkelley at hel4.arl.mil (Troy Kelley) Date: Wed, 10 Feb 1999 15:49:30 -0600 Subject: ACT-R model of spatial rotation Message-ID: Hello, I am attempting to program an ACT-R model of spatial rotation. I have some data where soldiers using a helmet mounted display do better on post tests of spatial rotation when compared to the pre tests of spatial rotation. Apparently, they are practicing some of their spatial rotation skills while using the helmet mounted display, so they end up doing better on the post test of spatial rotation. I am having a little trouble conceptualizing spatial rotation tasks as productions. I suppose I could have a group of "spatial skill productions" which could include compare productions, align productions, rotate X degrees productions and so on. Some of which may be used during the helmet mounted display tasks. These productions would then become stronger from repeated activation, and so the subjects would then perform better during the spatial rotation test using these same productions. Does this sound like a good way to proceed or should I use another strategy to represent this problem? Any advice would be appreciated. Troy From Todd.R.Johnson at uth.tmc.edu Fri Feb 12 19:25:15 1999 From: Todd.R.Johnson at uth.tmc.edu (Todd R Johnson) Date: Fri, 12 Feb 1999 18:25:15 -0600 Subject: List Memory and Alphabetic Retrieval Message-ID: As many of you know, I have been working on models of computation and recall in alphabetic retrieval (in the context of alphabet arithmetic). I'm wondering if any of you can shed any light on the following problem. In the original model I used a standard chunk-based representation for the alphabet: (alpha1 ISA item first a second b third c fourth d last-pos fourth next alpha2 parent alphabet) This means that if you want to know what comes after b, you must retrieve the chunk that b is in, then walk through the items until you reach b, then retrieve the next letter. This is consistent with Klahr's model of alphabetic retrieval and with the phenomenon that when we retrieve the chunk containing a probe, we must then start at the beginning of the chunk in order to locate the position of the probe. There are several problems with this representation. One is that it is completely inconsistent with that used in the Act-R models for serial and list memory. One might argue that since the alphabet is a well-learned list, it might deserve a different representation from lists such as those used in typical serial recall experiments. However, this difference seems somewhat troubling to me, and at best, is something that we should experimentally evaluate. A second problem is that backward recall is not much harder than forward recall. Since each chunk points only to the next chunk, it takes longer to retrieve the previous chunk, but backward recall within a chunk is just as easy as forward recall within a chunk. A third problem is that the data on alphabetic retrieval shows that the time to retrieve the initial letter of each chunk is a function of that chunk's position in the list. Several researchers have suggested that this means that people serially retrieve all of the chunks that are located prior to the chunk containing the probe letter. The contents of the prior chunks are not checked explicitly, but somehow people know to stop at the correct chunk. Nothing about the chunk-based representation enforces that kind of processing. It is a simple matter to retrieve the chunk containing the probe letter. There is no need to walk through the other chunks. The Act-R models for serial and list memory used positional encoding, such as: (alpha1 isa group list alphabet position first size 7) (atok isa token parent alpha1 position first name a list alphabet) (btok isa token parent alpha1 position second name b list alphabet) The serial recall model retrieves the group, then the elements in each group for forward recall. For backward recall the model retrieves each group from the first to the to-be-recalled group, then retrieves the elements of the that group. This is consistent with Klahr's model of alphabetic retrieval. However, I see no real need to do this, given the positional representation. The model can easily retrieve the last group (or any other group) directly. You might argue that the model would not know how long the list is, or how many groups, but the subjects were aware of this information. It is also unclear how to handle alphabetic retrieval given the positional representation. If you want to know what letter comes after m, it seems most obvious to simply recall the alphabet token containing m. This is consistent with the Act-R model of recognition memory. However, once this token is retrieved, it should be a simple matter to retrieve either the next token, or the previous token. In other words, there is no real need to step through all the groups to get to a later group. Nor is there any need to back up to the first element in a group. Suppose that we assume that the probe plus the alphabet list cue provide too little activation to successfully retrieve the correct token. Now it makes sense to step through each group, so that we can use the group name, plus the alphabet list, plus the probe as cues. So as each group is retrieved, the model would attempt to retrieve a token that is in that group and contains the probe. However, once we retrieve the correct token, there is still no need to then go back and retrieve the first token in the list. In otherwords, we can simply go directly to the next or previous letter. So it seems that neither representation directly requires the kind of access that appears to be in use in human alphabetic retrieval. One alternative is to use a purely associative representation of the alphabet list, but this approach also has problems. --- Todd Todd R. Johnson http://www.sahs.uth.tmc.edu/trjo hnso Associate Professor todd.r.johnson at uth.tmc.e du UT-Houston, School of Allied Health Sciences 713-500-3921 (voice) Dept. of Health Informatics 713-500-3929 (fax) "In the last 10 years we have come to realize humans are more like worms than we ever imagined," Dr. Bruce Alberts, president of the National Academy of Sciences. From dmcnamar at odu.edu Sat Feb 13 10:52:15 1999 From: dmcnamar at odu.edu (Danielle S. McNamara) Date: Sat, 13 Feb 1999 10:52:15 -0500 Subject: List Memory and Alphabetic Retrieval Message-ID: Todd, It seems to me that alphabetic retrieval is more analogous to skilled memory than to serial recall. That is, rather than relying on an episodic representation, the subject may be using a LTM (LTWM) representation. So, the process would be more similar to that observed in work such as Chase & Ericsson (S.F.) than in a standard serial recall paradigm. I say this because the subject would rely on a chunk in LTM, rather than a chunk created at the time of encoding. Also, the results you describe sound more similar to skilled memory findings. Does this assumption sound reasonable? And, has anyone simulated with ACT-R the Chase and Ericsson data -- or similar data? Danielle Danielle S. McNamara Asst. Professor Department of Psychology Old Dominion University Norfolk, Virginia 23529-0267 email: dmcnamar at odu.edu phone: (757) 683-4446 fax: (757) 683-5087 From ja+ at CMU.EDU Sat Feb 13 11:50:11 1999 From: ja+ at CMU.EDU (John Anderson) Date: Sat, 13 Feb 1999 11:50:11 -0500 (EST) Subject: List Memory and Alphabetic Retrieval Message-ID: There was an ACT-R model that dealt with the tasks of retrieving the next or the previous element in a list and a good bit of data Mike Matessa and I gathered about performing this task in new lists. One thing I remember is that we got the same pattern of results for new lists as Klahr et al got for the alphabet. In modeling this data we used, more or less, the current serial memory representation of one chunk per list element and organized these into chunks. That is, rather than representing the list Excerpts from mail: 12-Feb-99 List Memory and Alphabetic .. by "Todd R Johnson"@uth.tmc > (alpha1 ISA item first a second b third c fourth d last-pos fourth next > alpha2 parent alphabet) We represented it, more or less: Excerpts from mail: 12-Feb-99 List Memory and Alphabetic .. by "Todd R Johnson"@uth.tmc > (alpha1 isa group list alphabet position first size 7) > (atok isa token parent alpha1 position first name a list alphabet) > (btok isa token parent alpha1 position second name b list alphabet) There are a number of reasons for this representation. One is to enable positional confusions by partial matching of terms like first and second. Another is, as Todd notes, that it explains why it is longer to get to later items in the group. Please note (and I hope I keep it straight) that "chunk" refers to an ACT-R declarative structure and "group" to a set of elements in the serial list. I should note that this representational maneuver (breaking a chunk into a number of chunks, one for each part of the original chunk) is a frequent occurence in ACT-R models. For instance, Dario Salvucci has done this in his analogy model. There are a number of questions about this representation and Todd is to be thanked for hitting them on the head. I will address two of the deep ones: Todd raises concerns the issue of what forces recall to be forward-only given such a representation. My thoughts on this, such as they are, is that this has to do with incrementing the positional pointer --e.g., from first to second. The actual model uses LISP code to do this and my feeling is that the mind can only increment and not decrement positional indices. The motivation for this proposal is a bit fuzzy but there is observation that most of the time we search lists in a forward direction and so that is the direction that needs the support. The second serious question that Todd raises is Excerpts from mail: 12-Feb-99 List Memory and Alphabetic .. by "Todd R Johnson"@uth.tmc > It is also unclear how to handle alphabetic retrieval given the positional > representation. If you want to know what letter comes after m, it seems most > obvious to simply recall the alphabet token containing m. This is consistent > with the Act-R model of recognition memory. However, once this token is > retrieved, it should be a simple matter to retrieve either the next token, > or the previous token. The model with Mike Matessa did step through the list, first by groups and then by items within a group, as Todd describes and as the Klahr model did. My view, again such as it is, is an elaboration of the answer above. This is that one cannot really retrieve positional pointers but rather has to generate them. Thus, the only way to know what position an item occupies is to count up positional pointers until one gets to it. Indeed, I think we would concede we can only retrieve the position of m in its group by generating (abcd) (efg) (hijk) (lmnop) and saying second. Both points illustrate the fact that a significant piece of the ACT-R theory of serial memory does not come from the ACT-R architecture but rather assumptions about the position-based representation which are motivated by the data. From ja+ at CMU.EDU Sat Feb 13 12:05:07 1999 From: ja+ at CMU.EDU (John Anderson) Date: Sat, 13 Feb 1999 12:05:07 -0500 (EST) Subject: List Memory and Alphabetic Retrieval Message-ID: Excerpts from mail: 13-Feb-99 Re: List Memory and Alphabe.. by Danielle S. McNamara at odu > And, has anyone simulated with ACT-R the Chase and Ericsson data -- or > similar data? Peter Delaney and Jeff Feddon did start on such a model. I think it is a very important topic to pursue. From dmcnamar at odu.edu Sat Feb 13 12:33:30 1999 From: dmcnamar at odu.edu (Danielle S. McNamara) Date: Sat, 13 Feb 1999 12:33:30 -0500 Subject: List Memory and Alphabetic Retrieval Message-ID: John, Perhaps I have misunderstood, but when you said that we cannot retrieve positional pointers, I was reminded of some Estes data in STM serial recall. He looked at intrusions between lists and showed that if an item from a previous listed intruded into the recall of a subsequent list, the item retained its position in the list (or within 1 position). So, this indicates that the position of the item is not necessarily generated on line but is intrinsic to the encoding of the item. If I remember correctly, you and Mike simulated this type of data (i.e., item/order recall) -- would it have or did it predict these types of intrusions across lists? If so, how? Danielle > >The model with Mike Matessa did step through the list, first by groups >and then by items within a group, as Todd describes and as the Klahr >model did. My view, again such as it is, is an elaboration of the >answer above. This is that one cannot really retrieve positional >pointers but rather has to generate them. Thus, the only way to know >what position an item occupies is to count up positional pointers until >one gets to it. Indeed, I think we would concede we can only retrieve >the position of m in its group by generating (abcd) (efg) (hijk) (lmnop) >and saying second. > >Both points illustrate the fact that a significant piece of the ACT-R >theory of serial memory does not come from the ACT-R architecture but >rather assumptions about the position-based representation which are >motivated by the data. Danielle S. McNamara Asst. Professor Department of Psychology Old Dominion University Norfolk, Virginia 23529-0267 email: dmcnamar at odu.edu phone: (757) 683-4446 fax: (757) 683-5087 From klahr+ at andrew.cmu.edu Sat Feb 13 17:51:51 1999 From: klahr+ at andrew.cmu.edu (David Klahr) Date: Sat, 13 Feb 1999 17:51:51 -0500 (EST) Subject: List Memory and Alphabetic Retrieval Message-ID: A question and a pointer: The question: Has anyone run an experiment that attempts to combine the effects in the Klahr,Chase, Lovelace paper with the symbolic distance effects (i.e., the fact that the RT to decide which of two letters comes first is inversely proportional to their separation in the alphabet)? That is, has someone systematically looked at the SDE for pairs within and across chunk boundaries? It seems that such data would bear on the issues raised today. And if so, is there a single model that can account for the data? The pointer is just to note that there was a recent critique and a response to the KCL paper: Klahr, D.(1994) Plausible models of Alphabetic Search: A reply to Scharroo, Leeuwenberg, Stalmeier, & Vos (1994)Journal of Experimental Psychology: Learning, Memory, and Cognition. 20 (1). dk From niels at tcw2.ppsw.rug.nl Mon Feb 15 04:28:01 1999 From: niels at tcw2.ppsw.rug.nl (Niels Taatgen) Date: Mon, 15 Feb 1999 10:28:01 +0100 Subject: List Memory and Alphabetic retrieval Message-ID: Todd, The fact something goes faster in one direction than in the other direction, directional assymetry, is often quoted as evidence for production rules. So I feel the explanation has to be sought in the fact that something is proceduralized in one direction but not in another. The most simple solution is that people have rules like IF f THEN g, but not IF g THEN f. This simple solution is not entirely satisfactory, of course. In ACT-R, a rule needs to match a specific goal, so we cannot have a general purpose IF f THEN g rule. The answer should be more in line with John's idea that people are only accostumed to read the alphabet in one direction (or even lists in general). They might have some general purpose plan to read the alphabet, and use this plan along with other plans to do your task. If they have to read the alphabet backwards, they do not have a clear-cut plan, and have to create something new. This may be a not-yet-proceduralized (=declarative) plan of what to do, which is slow and prone to errors. Niels. -- ------------------------------------------------------------- Niels Taatgen Technische Cognitiewetenschap/Cognitive science & engineering Grote Kruisstraat 2/1, 9712 TS Groningen, Netherlands 050-3636435 / +31503636435 niels at tcw2.ppsw.rug.nl http://tcw2.ppsw.rug.nl/~niels ------------------------------------------------------------- From niels at tcw2.ppsw.rug.nl Mon Feb 15 06:04:56 1999 From: niels at tcw2.ppsw.rug.nl (Niels Taatgen) Date: Mon, 15 Feb 1999 12:04:56 +0100 Subject: Two ACT-R papers Message-ID: At the last ACT-R workshop, people agreed that it would be a good idea to announce new ACT-R papers on the mailing list. Unfortunately, this is hardly ever put into practice. Therefore, I want to announce two papers that Linda Jongman and I have submitted to the cognitive science conference. These papers can be retrieved from the web from http://tcw2.ppsw.rug.nl/prepublications/ An ACT-R model of individual differences in changes in adaptivity due to mental fatigue Linda Jongman and Niels Taatgen Abstract: In this paper we show that adaptivity is reduced when people become fatigued. Fatigued people adapt worse to changing probability distributions as compared to non-fatigued individuals. In an ACT-R model of the task we show that this decreased adaptivity is due to a decrease in the use of one specific strategy. We argue that the use of this strategy is decreased, because it places high demands on working memory. In previous research we also found indications that mental fatigue is related to changes in working memory functioning. We argue that modeling individual differences in performance will provide better insight in the processes involved in mental fatigue. A model of learning task-specific knowledge for a new task Niels Taatgen Abstract: In this paper I will present a detailed ACT-R model of how the task-specific knowledge for a new, complex task is learned. The model is capable of acquiring its knowledge through experience, using a declarative representation that is gradually compiled into a procedural representation. The model exhibits several characteristics that concur with Fitts theory of skill learning, and can be used to show that individual differences in working memory capacity initially have a large impact on performance, but that this impact diminished after sufficient experience. Some preliminary experimental data support these findings. -- ------------------------------------------------------------- Niels Taatgen Technische Cognitiewetenschap/Cognitive science & engineering Grote Kruisstraat 2/1, 9712 TS Groningen, Netherlands 050-3636435 / +31503636435 niels at tcw2.ppsw.rug.nl http://tcw2.ppsw.rug.nl/~niels ------------------------------------------------------------- From wolfgang.schoppek at uni-bayreuth.de Mon Feb 15 07:32:57 1999 From: wolfgang.schoppek at uni-bayreuth.de (Wolfgang Schoppek) Date: Mon, 15 Feb 1999 13:32:57 +0100 Subject: List Memory and Alphabetic Retrieval Message-ID: I think the alphabet is so highly overlearned that each letter is likely to be element of several chunks (even intraindividually). E.g. many letters could point to rather "fuzzy"chunks like "beginning of the alphabet" or "mid of the alphabet". That representation could explain the symbolic distance effect and can give a clue for entry points other than the beginning of the alphabet. -- Wolfgang -------------------------------------------------------------------- Dr. Wolfgang Schoppek <<< Tel.: +49 921 555003 <<< Lehrstuhl fuer Psychologie, Universitaet Bayreuth, 95440 Bayreuth http://www.uni-bayreuth.de/departments/psychologie/wolfgang.htm -------------------------------------------------------------------- From altmann at osf1.gmu.edu Mon Feb 15 17:46:14 1999 From: altmann at osf1.gmu.edu (ERIK M. ALTMANN) Date: Mon, 15 Feb 1999 17:46:14 -0500 (EST) Subject: Two more ACT-R papers Message-ID: Are at hfac.gmu.edu/people/altmann/cogsci99.sea. The first presents a closed-form model of serial attention that integrates base-level and associative activation, showing that both are necessary to achieve empirical performance accuracy given noise in memory. The second presents a model of Tower of Hanoi data that stores goals in memory instead of on the goal stack. The model predicts average number of moves per trial, in addition to fitting latencies on optimal trials with R^2 = .99. Comments appreciated. Erik. Altmann E. M. & Gray, W. D. (1999). Serial attention as strategic memory. Abstract - Serial attention involves focussing on a sequence of thoughts under deliberate control. As such, it has two phases, attention switching and attention maintenance. Attention switching involves rapidly building up the activation of the new thought so it can be temporarily dominant. Attention maintenance allows the current thought to gradually decay to prevent it from intruding on the next. SASM, a model based on this analysis, suggests that this balance of initial activation followed by gradual decay reflects a strategic adaptation to architectural constraints and task demands. SASM makes accurate predictions about error patterns and encoding time, and integrates attention maintenance and attention switching in one unified theory. Altmann E. M. & Trafton, J. G. (1999). Memory for goals: An architectural perspective. Abstract - The notion that memory for goals is organized as a stack persists as a central feature of cognitive theory, in that stacks are primitive mechanisms in leading cognitive architectures. However, the stack construct over-predicts the strength of goal memory and the precision of goal selection order, while under-predicting the maintenance cost of both. A better approach to understanding cognitive goal management is to treat goals like any other kind of memory element. This makes accurate predictions and reveals the nature of goal encoding and retrieval processes in detail. The approach is demonstrated in an ACT-R model of human performance on a prototypically goal-based task, the Tower of Hanoi. The model and various theoretical and methodological considerations suggest that cognitive architectures should enforce a two-element constraint on the depth of the stack, to deter its use for storing task goals while preserving its use for basic cognitive processes like focussing attention and symbolic learning. From gray at gmu.edu Fri Feb 19 13:11:32 1999 From: gray at gmu.edu (Wayne Gray) Date: Fri, 19 Feb 1999 13:11:32 -0500 Subject: Free GOMS Tutorial Message-ID: WHAT This full-day tutorial introduces systems design from the perspective of cognitive process analysis and modeling. It describes cognitive issues involved in user interface design, as well as the benefits of using various types of cognitive modeling tools (GOMS, NGOMSL, CPM-GOMS) to describe and predict human performance in using complex, dynamic systems. The tutorial will include examples drawn from industry as well as hands-on practice in applying the techniques learned. WHEN: Monday, May 17th 1999. The day before the CHI'99 Technical Program begins. WHEN: 9 AM to 4 PM WHERE: Carnegie Mellon University, Pittsburgh, PA MORE INFO: http://hfac.gmu.edu/free-goms _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ Wayne D. Gray HUMAN FACTORS & APPLIED COGNITIVE PROGRAM SNAIL-MAIL ADDRESS (FedX et al) VOICE: +1 (703) 993-1357 George Mason University FAX: +1 (703) 993-1330 ARCH Lab/HFAC Program ********************* MSN 3f5 * Work is infinite, * Fairfax, VA 22030-4444 * time is finite, * http://hfac.gmu.edu * plan accordingly. * _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ From psyee at showme.missouri.edu Wed Feb 24 21:26:34 1999 From: psyee at showme.missouri.edu (psyee at showme.missouri.edu) Date: Wed, 24 Feb 1999 20:26:34 -0600 (CST) Subject: spoken recall Message-ID: I was interested in finding out about how spoken recall could be implemented into the model. Has this been done? Thanks, Emily Elliott ******************************* Emily Elliott University of Missouri-Columbia ph. (573)882-7417 psyee at showme.missouri.edu www.missouri.edu/~psyee ******************************* From tkelley at hel4.arl.mil Thu Feb 25 12:36:30 1999 From: tkelley at hel4.arl.mil (Troy Kelley) Date: Thu, 25 Feb 1999 11:36:30 -0600 Subject: Comments vanishing?? Message-ID: Group, When I opened my ACT-R project this morning, all the comments I had put in the file the day before were gone. I am commenting statements like this ;;; here is a comment Any ideas as to what could have gone wrong? Troy From cl at andrew.cmu.edu Thu Feb 25 13:55:57 1999 From: cl at andrew.cmu.edu (Christian Lebiere) Date: Thu, 25 Feb 1999 13:55:57 -0500 Subject: Comments vanishing?? Message-ID: The comment monster ate them ;-). Seriously, for future general reference bug reports should be sent to act-r-users-request+ at andrew.cmu.edu rather than act-r-users+ at andrew.cmu.edu. It is also highly useful to provide details such as which version of ACT-R you are using (e.g. version number as printed in the startup screen, Mac or PC version of the environment, etc) and if possible a piece of the model (code, trace, etc) replicating the problem or at least a precise description of the circumstances. Thank you all, Christian --On Thu, Feb 25, 1999 11:36 AM -0600 Troy Kelley wrote: > Group, > > When I opened my ACT-R project this morning, all the comments I had put > in the file the day before were gone. I am commenting statements like > this > > ;;; here is a comment > > Any ideas as to what could have gone wrong? > > Troy >