Difference between revisions of "Explainer moves"

From INF Wiki
Jump to navigation Jump to search
 
(6 intermediate revisions by the same user not shown)
Line 1: Line 1:
<h2>Affect and motivation</h2>
A good teacher/tutor bolsters student motivation, confidence, and self-efficacy while mastering the material. Accordingly, the tutors should refraining from confronting, embarrassing, and humiliating  the  student. And polite conversationalists move towards affect and motivation by following politeness principles.<ref name="graesser1995collaborative">Graesser, A. C., Person, N. K., & Magliano, J. P. (1995). Collaborative dialogue patterns in naturalistic one‐to‐one tutoring. Applied cognitive psychology, 9(6), 495-522.‏</ref>
-'''Explainee:''' ["I'm in my first year of a PhD in Computer Science", "and I'm studying natural language processing", 'and machine learning.']
-'''Explainer:''' [ So '''''would you mind''''' telling me a bit about what you've been working on or interested in lately?] ---> '''''politeness indicates an affect and motivation move here'''''
-'''Explainee:''' ["I've been looking at understanding persuasion", 'in online text and the ways that we might be able to', 'automatically detect the intent behind that persuasion', "or who it's targeted at", 'and what makes effective persuasive techniques.']
-'''Explainer:''' ["So what are some of the techniques you're applying", 'to look at that debate data?']
<h2>Diagnostic query</h2>
<h2>Diagnostic query</h2>
An utterance by which ”the speaker is testing whether a listener knows a piece of information by asking them to supply the information”. The speaker, normally the tutor, already knows the answer. They are often questions, but can be also requests: ”tell me how electricity flows through the circuit” <!--\parencite{karagjosova2005dialogue, jang2014diagnostic}--><ref name="karagjosova2005dialogue">Karagjosova, E., & Tsovaltzi, D. (2005). Dialogue moves for DIALOG.</ref><ref name="jang2014diagnostic">Jang, E. E., & Wagner, M. (2014). Diagnostic feedback in the classroom. The companion to language assessment, 2, 157-175.‏</ref>
An utterance by which ”the speaker is testing whether a listener knows a piece of information by asking them to supply the information”. The speaker, normally the tutor, already knows the answer. They are often questions, but can be also requests: ”tell me how electricity flows through the circuit” <!--\parencite{karagjosova2005dialogue, jang2014diagnostic}--><ref name="karagjosova2005dialogue">Karagjosova, E., & Tsovaltzi, D. (2005). Dialogue moves for DIALOG.</ref><ref name="jang2014diagnostic">Jang, E. E., & Wagner, M. (2014). Diagnostic feedback in the classroom. The companion to language assessment, 2, 157-175.‏</ref>
Line 21: Line 11:
  -'''Explainee:''' ['I never knew beforehand how hard it was', 'to get actual data of the black holes itself,', "first of all, they're dark, and, like,", "they're so far away, it's almost impossible just", 'to get a good image of them.', 'They were discussing a project in', 'which multiple radio telescopes of some sort, like, are,', 'like, pinpointed all across, from Greenland', "to South America, and, like, and they're trying to", 'get an image of', 'the black hole in the center', 'of our galaxy because, as opposed', 'to just recording its impact', 'on the surrounding stars and planets.']
  -'''Explainee:''' ['I never knew beforehand how hard it was', 'to get actual data of the black holes itself,', "first of all, they're dark, and, like,", "they're so far away, it's almost impossible just", 'to get a good image of them.', 'They were discussing a project in', 'which multiple radio telescopes of some sort, like, are,', 'like, pinpointed all across, from Greenland', "to South America, and, like, and they're trying to", 'get an image of', 'the black hole in the center', 'of our galaxy because, as opposed', 'to just recording its impact', 'on the surrounding stars and planets.']
            
            
<h2>Elaborating</h2>
This is an assertion by the tutor that fills in missing information that the tutor regards as important, but otherwise might be missed by the student. In essence, information is simply transmitted from the tutor to the learner, as opposed to having the learner generate the information
<!--\parencite{GRAESSER199935}. --><ref name="GRAESSER199935">Graesser, A. C., Wiemer-Hastings, K., Wiemer-Hastings, P., Kreuz, R., & Tutoring Research Group. (1999). AutoTutor: A simulation of a human tutor. Cognitive Systems Research, 1(1), 35-51.‏</ref>
-'''Explainer:''' ["We're gonna talk about blockchain technology.", 'Have you heard of blockchain?']
-'''Explainee:''' ["I've heard of the words blockchain,", "but I'm not sure I know what it is."]
-'''Explainer:''' ['When we were much smaller societies,', 'you and I could trade in our community pretty easily.', 'As the distance in our trade grew,', 'we ended up inventing institutions, right?', 'If you Uber or you use Airbnb or you use Amazon even,', 'these are just digital marketplaces and platforms', 'that help us facilitate an exchange of value.', 'But today, we actually have a technology', 'that allows us to trade one to one but at scale,', "and it's called blockchain technology.", 'There is some kind of interface for it.', 'You could have an app, or you could use a computer', 'to do it, but instead of there being a company', "in the middle that's helping you make that transaction,", "it's a bunch of software code."]
-'''Explainee:''' ['Okay.']
-'''Explainer:''' ['''''And so it's being run by all of these different computers that have like a node. So they're all running the same software. and guaranteeing your transactions as they happen.''''']  --->'''''Elaborating'''''
<h2>Elicited responses</h2>
elicited responses would be a type of constructive and interactive ones in which the tutors make a move, such as asking a question, and the students respond to such elicitation. For example, if a tutor sets a goal or provides a procedural hint (for a problem solving dialogue), and a student responds by taking a step toward reaching that goal or elaborating upon that hint, then such a dialogue would be both constructive and interactive <!--\parencite{CHI2001471}-->.<ref name="CHI2001471">Chi, M. T., Siler, S. A., Jeong, H., Yamauchi, T., & Hausmann, R. G. (2001). Learning from human tutoring. Cognitive science, 25(4), 471-533.</ref>
<h2>Feedback, error diagnosis, and remediation</h2>
A good teacher/tutor should quickly give feedback on the quality of student contributions. When a student makes an error, the teacher/tutor should identify the error, correct the error, diagnose the misconception that explains the error, and rectify the misconception.<ref name="graesser1995collaborative"/>




Line 59: Line 32:
<h2>Hinting</h2>
<h2>Hinting</h2>
Hinting is an explanation move and a type of scaffolding where something is suggested or indicated indirectly or covertly. Examples: "The hard disk can be used for storage" or “What about the hard disk?”  
Hinting is an explanation move and a type of scaffolding where something is suggested or indicated indirectly or covertly. Examples: "The hard disk can be used for storage" or “What about the hard disk?”  
"When the student is having problems answering a question or solving a problem, the tutor gives hints by presenting a fact, asking a leading question, or reframing the problem. A hint may be a memory cue or a critical problem-solving clue. Hints are frequently indirect speech acts, so they run the risk of being missed by an insensitive student" <!--\parencite{GRAESSER199935}-->.<ref name="GRAESSER199935"/>
"When the student is having problems answering a question or solving a problem, the tutor gives hints by presenting a fact, asking a leading question, or reframing the problem. A hint may be a memory cue or a critical problem-solving clue. Hints are frequently indirect speech acts, so they run the risk of being missed by an insensitive student" <!--\parencite{GRAESSER199935}-->.<ref name="GRAESSER199935">Graesser, A. C., Wiemer-Hastings, K., Wiemer-Hastings, P., Kreuz, R., & Tutoring Research Group. (1999). AutoTutor: A simulation of a human tutor. Cognitive Systems Research, 1(1), 35-51.‏</ref>
 
<h2>Immediate feedback</h2>
Graesser documented some of the dialog moves that tutors generate to nurture the collaborative building of explanations. There is positive, neutral, and negative feedback, as illustrated below <!--\parencite{GRAESSER199935, person:hal-00197320}. --><ref name="GRAESSER199935"/><ref name="person:hal-00197320">Person, N. K., Graesser, A. C., Kreuz, R. J., & Pomeroy, V. (2003). Simulating human tutor dialog moves in AutoTutor. International Journal of Artificial Intelligence in Education (IJAIED), 12, 23-39.‏</ref>
  (1) Positive immediate feedback. ‘That’s right,’ ‘Yeah.’
  (2) Neutral immediate feedback. ‘Okay,’ ‘Uh- huh.’
  (3) Negative immediate feedback. ‘Not quite,’ ‘No.’ 
 
After this immediate feedback, the tutor normally issues a more substantive dialogue move such as pumping and prompting.
 
'''Example (1):'''
-'''Explainee:''' ['Music is not different from life.']
-'''Explainer:''' ['''''No.'''''] ---> '''''Negative immediate feedback'''''
-'''Explainee:''' ["You know, and I think that's probably", 'the greatest attraction to those of us who play music.']
-'''Explainer:''' ['''''Yeah.''''', 'Was there ever a point in your life', 'when you were younger where you felt like you had', 'consistently fell back into the same habits?'] ---> '''''Positive immediate feedback'''''
 
<!--<h2>Introducing extraneous information</h2>
Introducing extraneous information is an explanation move where an explanation is enhanced by providing extraneous information and a more complete and coherent understanding.-->
 
<h2>Open option</h2>
An utterance that suggests a course of action or states a possibility. Imperatives are normally not open options.<!--\parencite{allen1997draft}--> <ref name="allen1997draft">Allen, J., & Core, M. (1997). Draft of DAMSL: Dialog act markup in several layers.‏</ref>
 
<h2>Prompting</h2>
"The tutor supplies the student with
a discourse context and prompts him / her to fill in a
missing word, phrase, or sentence. An example in
the context of computer literacy would be ‘The
primary memories of the CPU are ROM and... .’
The prompt is delivered with intonation, facial expressions, and gestures that signal the learner to fill in the missing word or phrase. The one or two words that precede the missing word are drawn out, with a complex, bending intonation contour. Then there is a pause that gives the floor to the learner and invites the learner to fill in the missing information. Sometimes the facial expression or hand gestures have an encouraging stance, as if to say, ‘Give me the next word.’ Prompting is a scaffolding device for students who are reluctant to supply information. Students are expected to supply more content and more difficult content as they progress in learning the domain knowledge." <!--\parencite{GRAESSER199935, person:hal-00197320}.--> <ref name="GRAESSER199935"/><ref name="person:hal-00197320"/>
 
Karagjosova and Tsovaltzi <!-- \parencite{karagjosova2005dialogue} --><ref name="karagjosova2005dialogue"/> define prompt as an utterance that requests the hearer explicitly to proceed with the task and provide further information. Examples of such a move are as follows:
 
  Examples:
    -What's next?
    -How could that work?
    -How could it go on?
 
 


<h2>Pumping</h2>
<h2>Pumping</h2>
Line 123: Line 59:


   Example (2): Embedded correction <ref name="clark_bernicot_2008"/>:
   Example (2): Embedded correction <ref name="clark_bernicot_2008"/>:
         Child: He falled, he falled again.
         '''Child:''' He falled, he falled again.
         Father: Ok he fell, but no, he’s at the boat, now put him in front of the car
         '''Father:''' Ok he fell, but no, he’s at the boat, now put him in front of the car
          
          




Line 134: Line 68:


"The tutor jumps in and splices correct information as soon as the student produces a contribution that is obviously error-ridden. The tutor needs to be able to recognize errors, bugs, and slips in order to do this. Deep misconceptions in the student are more difficult to detect and are not handled by splicing."  
"The tutor jumps in and splices correct information as soon as the student produces a contribution that is obviously error-ridden. The tutor needs to be able to recognize errors, bugs, and slips in order to do this. Deep misconceptions in the student are more difficult to detect and are not handled by splicing."  
<!--\parencite{GRAESSER199935, person:hal-00197320} --><ref name="GRAESSER199935"/><ref name="person:hal-00197320"/>
<!--\parencite{GRAESSER199935, person:hal-00197320} --><ref name="GRAESSER199935"/><<ref name="person:hal-00197320">Person, N. K., Graesser, A. C., Kreuz, R. J., & Pomeroy, V. (2003). Simulating human tutor dialog moves in AutoTutor. International Journal of Artificial Intelligence in Education (IJAIED), 12, 23-39.‏</ref>
    
    
<h2> Requestioning</h2>
The tutor re-asks the original main question when the thread of the tutorial dialog is moving off course or on a tangent
<!--\parencite{GRAESSER199935}.--><ref name="GRAESSER199935"/>
<h2>Understanding query</h2>
An utterance that asks “the listener whether they understood, rather than making them prove it. It takes a “yes” or “no” as an answer  <!--\parencite{karagjosova2005dialogue}. --><ref name="karagjosova2005dialogue"/>
Difference to diagnostic query: the tutor does not know the answer. The realisation of the two can be the same (”Do you understand?”), but the obligations that they pose are different: diagnostic queries require more than ”yes” or ”no” as an answer.
   
<!--Difference to info request: It asks for signalling understanding.-->
'''Example (1):'''
-Do you now understand how this relates to the implication?
==Notes==
==Notes==
<references />
<references />

Latest revision as of 22:39, 27 September 2022

Diagnostic query

An utterance by which ”the speaker is testing whether a listener knows a piece of information by asking them to supply the information”. The speaker, normally the tutor, already knows the answer. They are often questions, but can be also requests: ”tell me how electricity flows through the circuit” [1][2]

Example (1):

-Explainer: [So tell me, what do you know about black holes?] --->Diagnostic query
-Explainee: ["Well, I know they're created when stars, once they start growing, it doesn't, it can't expand anymore, so they collapse inward.]
-Explainer: ['You got, pretty much, you got a very good idea about it...]

Example (2):

-Explainer: [So far, what do you know about black holes?]  --->Diagnostic query
-Explainee: ['I never knew beforehand how hard it was', 'to get actual data of the black holes itself,', "first of all, they're dark, and, like,", "they're so far away, it's almost impossible just", 'to get a good image of them.', 'They were discussing a project in', 'which multiple radio telescopes of some sort, like, are,', 'like, pinpointed all across, from Greenland', "to South America, and, like, and they're trying to", 'get an image of', 'the black hole in the center', 'of our galaxy because, as opposed', 'to just recording its impact', 'on the surrounding stars and planets.']
          


Generating a recast

Clarke et. al. [3] discuss that "in the adult–child interaction, the adult generates a recast when he or she "....expands, deletes, permutes, or otherwise changes the [child’s utterance] while maintaining significant overlap in meaning". Sometimes, recasts provide a corrective contrast with the child’s immediately prior turn through the provision of an enhanced version of the child’s ill-formed utterance, as in the example below reproduced from Saxton."

    "Child: It might get loosed down the plughole
    Adult: Lost down the plughole?"

"Equally, some recasts give no clear-cut correction but still offer potential for facilitation of acquisition by providing a structural contrast with an errorless sentence produced by the child as in the following constructed extract, which illustrates an adult expanding the child’s utterance into one that is more complex"[3]:

   "Child: That's a big horse. 
   Adult: Yep, that's a big horse with lots of black spots."

"Sometimes, recasts may introduce multiple changes, as in the following example,whereby the adult both expands on and corrects components of the child’s utterance:"[3]

   "Child: Why does he that? 
   Adult: Mmm, I am not sure why he does that." 


Hinting

Hinting is an explanation move and a type of scaffolding where something is suggested or indicated indirectly or covertly. Examples: "The hard disk can be used for storage" or “What about the hard disk?” "When the student is having problems answering a question or solving a problem, the tutor gives hints by presenting a fact, asking a leading question, or reframing the problem. A hint may be a memory cue or a critical problem-solving clue. Hints are frequently indirect speech acts, so they run the risk of being missed by an insensitive student" .[4]

Pumping

Pumping is a scaffolding mechanism which is used if a tutor wants the student to contribute more information in the dialogue and thereby be more active. The tutor pumps the student to elaborate an answer by giving positive or neutral feedback (e.g. ‘uh huh’, ‘okay’, head nod) and then pausing for the student to supply more information.

"The tutor pumps the student for more information during the early stages of answering a particular question (or solving a problem). The pump consists of positive feedback (e.g., ‘Right,’ ‘Yeah,’ dramatic head-nod), neutral backchannel feedback (‘Uh-huh,’ ‘Okay,’ subtle head-nod), or explicit requests for more information (‘Tell me more,’ ‘What else?’). The tutor pumps for one or two cycles of turns before the tutor contributes information. Pumping serves the functions of exposing knowledge of the student and of encouraging students to construct content by themselves." [4][5]

Example (1):

-Explainer: ["So essentially, the way I'd apply negative harmony", 'would be this idea of polarity, you know,', 'between the overtone series and the undertone series,', 'or you know, the one side and the other side.', 'The perfect and the play goal.', 'The feeling of a minor perfect--', '(keyboard music)', "Resolving, it's so moving, you know?", "And it's a good alternative to something like,", '(keyboard music)']
-Explainee: ["It's funny, you know, you doing that", 'makes something in a major key', 'sound like kind of a wistful sad song.']
-Explainer: [Right.] ----> Pumping
-Explainee: ['You know, you changed the feeling of it,', 'what otherwise would, you know, if you were to tell a kid', 'that this is a major song, we should be happy.']

repetition to place information in common ground, reformulation, Embedded correction

Repetition has multiple functions in conversation [6]. One pervasive use, for example, is when a speaker adds to common ground new information offered by another.

"As an example in child-adult interaction, some adult repetitions are characterized as ‘reformulations’. These are repetitions of erroneous child utterances in which adult speakers have corrected the child errors, underlined, as in example (1)"[6]:

Example(1)[6]:
     Child: Don ’t fall me downstairs!
     Parent: oh, I wouldn't drop you downstairs.
     Child: Don ’t drop me downstairs.


"Adults also make some embedded corrections, as in Example (2), where the correction of the erroneous form is (fell) offered in the next turn by the father"[6]:

 Example (2): Embedded correction [6]:
       Child: He falled, he falled again.
       Father: Ok he fell, but no, he’s at the boat, now put him in front of the car
       


splicing

splicing is considered as a scaffolding strategy. When a student’s contribution is error-ridden, the tutors tend to jump in quickly and splice in a good answer. Thus, immediate corrective feedback is given by the tutor. The metaphor of ‘splicing’ is used because normally the tutor and student are jointly constructing a connected structure of ideas when the errors occur.

"The tutor jumps in and splices correct information as soon as the student produces a contribution that is obviously error-ridden. The tutor needs to be able to recognize errors, bugs, and slips in order to do this. Deep misconceptions in the student are more difficult to detect and are not handled by splicing." [4]<[5]

Notes

  1. Karagjosova, E., & Tsovaltzi, D. (2005). Dialogue moves for DIALOG.
  2. Jang, E. E., & Wagner, M. (2014). Diagnostic feedback in the classroom. The companion to language assessment, 2, 157-175.‏
  3. 3.0 3.1 3.2 Clarke, M. T., Soto, G., & Nelson, K. (2017). Language learning, recasts, and interaction involving AAC: Background and potential for intervention. Augmentative and Alternative Communication, 33(1), 42-50.‏‏
  4. 4.0 4.1 4.2 Graesser, A. C., Wiemer-Hastings, K., Wiemer-Hastings, P., Kreuz, R., & Tutoring Research Group. (1999). AutoTutor: A simulation of a human tutor. Cognitive Systems Research, 1(1), 35-51.‏
  5. 5.0 5.1 Person, N. K., Graesser, A. C., Kreuz, R. J., & Pomeroy, V. (2003). Simulating human tutor dialog moves in AutoTutor. International Journal of Artificial Intelligence in Education (IJAIED), 12, 23-39.‏
  6. 6.0 6.1 6.2 6.3 6.4 Clark, E. V., & Bernicot, J. (2008). Repetition as ratification: How parents and children place information in common ground. Journal of child language, 35(2), 349-371.‏