11/19/2009

ASD social problems: a failure of common ground?

I've been thinking a lot lately about the implications of "common ground."

The concept of "common ground" was so obvious that it wasn't defined and studied much until recently. (It's currently a popular organizing principle in cognitive psychology research on communication, itself a fairly new area). The basic idea is that when we speak to each other, we need a common reference to talk about. If neither person has knowledge about the world in common with the other, then they will be unable to communicate. A conversation might go like this:

American college student #1: "Where are you from?"
American college student #2: "A stone age tribe in Indonesia."
ACS 1: ::pauses, unsure what to ask:: "...What was that like?"
ACS 2: ::how on earth do you answer a question like that when nothing in a stone age tribe in Indonesia is familiar to an American college student?:: "Uh...I dunno."
ACS 1: "Oh."

If there were common ground, the conversation would be more like this:
American college student #1: "Where are you from?"
American college student #2: "I'm from Miami."
A.C.S. 1: "Oh yeah, my cousin's from there. Do you miss all the parties?"
A.C.S. 2: "Nah, I kinda like the quiet here. But I do miss the hot weather!"
A.C.S. 1: "Yeah, Minnesota weather is the worst. Do you know how deep the snow got last year?"
...etc.

This is why people always ask the same boring questions: "what are you majoring in?" "Where are you from?" "Do you know so and so?" If I know something about your major, or know someone who has your major, or had a teacher who taught classes in your major, I can ask you questions about it and share my experiences. Furthermore, each topic opens up a rich vein of potential conversational topics. If I know nothing about it, I can't think of anything to ask, and I can't share any relevant experiences.

As banal as these observations seem, they have an interesting implication: we might associate with people of our own race and socioeconomic background, not because we're prejudiced, but because we have more common ground that way. It's easier to talk to people with whom we share a lot of experiences, and we tend to do what's easy.

The concept of common ground might also be a reason why adolescents don't talk to their parents about their lives (although I'd imagine privacy, not wanting to be judged, or wanting to separate from their parents all play a role, too). If you think you live in a different world than your parents did, if you think your parents don't remember what it's like to be your age, or if you think your parents won't understand what you and your friends talk about, you're likely to think you don't have much to say to your parents.

Most studies of common ground involve one stranger teaching another how to do something. They demonstrate how opaque speech can be and how much communication relies on two people seeing the same thing:

Instructor: "First you, um, take this thing here--"
Doer: "The thing to the left of the red one?"
Instructor: "No no no--yeah, yeah, that's it--"
Doer: "Do I put it here?"
Instructor: "No, to the right of--yeah, to the right of that hole...you got it...now turn it over?"
Doer: "Wait, do I turn this over or do I turn that over?"
Instructor: "The bigger one."
Doer: "O.K."

Or, as a friend once said to me: "it's the thingy between the thingy and the other thingy." She was serious.

Conversational partners develop a set of labels that they use repeatedly with the same partner. They'll start with a long, descriptive label--"the blue bird with the red belly and the pointy beak to the left of the cardinal" and will gradually shorten it to "the blue and red bird" or "the blue pointy bird" and will use that label every time with the same partner. When they do the same task with a new partner, they may develop a different label; say, the "blue robin" instead of the "blue pointy bird."

For the labeling process to work, and conversational partners to agree on what objects are present, both partners must see the same features: the bird's blueness, the redness of its belly, and the pointiness of its beak, for example. They must also realize that the other person sees the same features they do--in other words, that their conversation partner's "blue and red bird" is the same as your "blue and red bird," and your conversation partner knows this. The complexity of this simple, everyday task leaves a lot of stumbling blocks for people with learning disabilities.

The first problem is that conversational partners might not see the same features in the first place. Consider the problems that arise when an experienced computer user is trying to teach an elderly person who has never used a computer before:

::they've just discussed how to turn the computer on and are getting down to the business of using it::
Young person: "Put your hand on the mouse."
Old person: "What's the mouse?"
YP: "You know, the gray thing that you click on."
OP: "..." ::Looks between the keyboard, which has buttons you click on, the mouse, which has buttons, and the computer, which after all has an on button you can click. All are gray.::
YP: "It has buttons."
OP: "...This?" ::picks up the mouse::
YP: "Yes. Now scroll to the start menu, left click on it, and hold the mouse button there."
OP: "What do you mean by scroll? What does the mouse do? What is left clicking? What am I trying to do?"
YP: "You're trying move the mouse around the screen so you can open a program."
OP: ::picks up the mouse and holds it on the surface of the screen:: "I put the mouse on the screen, why isn't it doing anything?"
YP: "No! You don't do that! It's a mouse, how do you not get this? I give up."

Yes, there really are people who don't know that a mouse is supposed to move an arrow around the screen that lets you interact with icons that represent files stored on the computer. And if you don't explain that, they certainly won't understand any fancy terms you throw around like "open a new tab in your browser" or "now you're going to work in an environment." And "directories?" Forget it.

In these situations, the old person rarely learns much about how to use a computer. The old person feels condescended to while the young person feels frustrated and unappreciated. If situations like this happen enough to the old person, he or she may just decide this newfangled technology stuff is too complicated and give up on trying to learn it entirely. Notice any parallels with learning disabilities?

The defining feature of a learning disability, for many kids and their parents, is the need to verbally explain a concept that everyone just knows. For instance, everyone just knows that you don't walk up to random strangers and grope them, but I know a boy with ASD who has to be told over and over. In an earlier entry, I described a dyspraxic person who viewed a coffeemaker as like a black box and did not understand how it worked. These errors are frustrating to neurotypicals because the correct behaviors are too obvious to be formulated s a rule or even verbalized. Experts in any domain are bad at explaining how they do what they do, but it's especially bad in practical and social domains. This spells trouble for people with learning disabilities that affect their practical and social functioning: they desperately need explained what only trained professionals can explain, just so they can share some common referents with the rest of us.

If sensory processing problems are severe enough at an early enough age, they seem to prevent kids from forming basic academic, practical, or social concepts, thus leading to the academic, practical, or social blindness that so frustrates neurotypicals. But this doesn't always happen; after all, some people's sensory processing problems go underground until adolescence or adulthood. However, these processing differences can still affect common ground. Such a person may not notice subtle features that neurotypicals notice, or they may notice a different set of features.

Once they have referents, people need to know that their conversational partner shares them. This is difficult for people with ASD for two reasons. First, people with ASD may be used to perceiving the world differently than everyone else. As a result, they correctly believe that others do not share their frame of reference--there is no common ground. Second, people with ASD are less able to use the compensating behaviors neurotypicals use to create common ground in difficult situations.

It turns out common ground can be difficult to establish even for neurotypicals (think about the conversation between the instructor and the doer or the old person and the young person). Explainers compensate through gesture: they use pointing and eye gaze to indicate an object. Comprehenders ask questions. For example:

Explainer: "The screwdriver you're looking for is in the study."
Comprehender: "Where in the study?"
Explainer: "It's over there" ::points at a table in the back of the room. However, pointing gestures can be a little vague sometimes, especially since there are several pieces of furniture in the back of the room. So the comprehender asks for clarification::
Comprehender: Do you mean that table? ::walks closer to it and points at it::
Explainer: Yeah, you got it. So it'll be uh, on the right side, behind, you know, the thing that holds the papers... ::moves hands to the right and mimes pushing it aside::
Comprehender: I just found a paperweight on the right there, and all I'm seeing behind it is a giant pile of papers. Are you sure it's there?
Explainer: Oh, sorry, I forgot I put it away. Actually, it's in the drawer...

People with ASD are bad at following gaze from infancy. They may not realize that gaze direction carries information. They also tend to be bad at asking questions. When Adam was in speech and language therapy, he had to be taught to ask his conversational partner questions, not just talk about his own thoughts. It took him a long time to internalize this. It also did not occur to him to ask questions when he didn't understand something someone said, which happened often when he was in early elementary school. Thus, he would often repeat phrases he heard without knowing what they meant--and he didn't know what they meant because he never asked.

In short: we can look at social deficits in ASD (and probably nonverbal learning disabilities) as a failure of common ground. Seems like a nice alternative to the "no theory of mind" explanation that doesn't work for a lot of high-functioning kids.