Monday, July 12, 2010

When Words Get in the Way - Verbal Mediation Interferes with Spatial Imagery

A colleague recently mentioned the work of Jonathan Schooler who discovered that talking about a problem interfered with problem solving by insight. Verbal mediation does not worsen all types of problem solving of course - in older adults, in fact, researchers found that having older adults talk aloud while solving Raven's matrices increased their IQ performance by 11 points.

But verbal overshadowing is a real phenomenon, and another example of counteracting processes of words and images in the brain.

Verbal overshadowing refers to the process by which putting experiences into words can certain aspects of visual and other perceptual memories.


-  After viewing a video of a bank robbery, subjects were 64% accurate picking the robber from a line if they relied on vision alone, but only 38% accurate if they gave a verbal description of the robber before searching the lineup.
- After viewing a map with landmarks, only those subjects who had been asked to verbally describe the map had trouble estimating the distances between landmarks. The thought is that words interfered with the spatial image of the map.
- After listening to a spoken phrase,  test subjects who were asked to write down everything they heard had a harder time identifying the voice that was used

More recent information about the verbal mediation suggests that individuals with stronger spatial ability are more susceptible to the verbal overshadowing that means that spatial thinkers may really not want to solve problems with words because it may disrupt their spatial processing and cause them to make more mistakes. In most of the students we see in the clinic where verbal mediation helps, their imagery for the task is weak (like dyslexic students verbalizing spelling words) - so this all makes sense.

In the figure at left, Melcher (2004) found that perceptual training increased the verbal overshadowing effect (in this experiment, subjects were trained on visual discrimination of different mushroom types).

Take-home point for teachers: when we're teaching or demonstrating a lesson, we should be attuned to primary modality we're trying to entrain. Sometimes we need to show and talk less.

Also if our students are spatial experts, we should understand why they don't like to "show their work" and not have it required of them.

Verbal Overshadowing of Spatial Mental Models
Verbal Overshadowing Articles
Talk picture


  1. Wow, this is really fascinating. I can't help but wonder how this applies to my son who has a PDD-NOS diagnosis ... he is very very good with puzzles beyond his age, but his language is definitely off. but which came first? it seems like maybe there could be a chicken and egg scenario with some of this stuff ... ?

  2. This post is definitely foder for possibly adjusting my 8 yo DS's Advanced Learner Plan (ALP). BTW, he's GT in Visual Spatial and Math; and has HFA (i.e., 2e HFA). His 2nd grade teacher complained this last year that he doesn't show his math work.

    So her major idea, if he's got extra time because of testing out of a math unit, was to learn to "Show his Work", possibly by keeping a math journal. Nothing like having a stick for a goal.

    Whereas, I suggested if he tests out of a unit, that he should be able to work on STEM (Science, Technology, Engineering, or Math) related web sites doing some applied problem solving scenarios that they've included on the site. I.e., work on developing his Visual Spatial by solving practical "hands on" tasks. More of a carrot goal approach, I hope.

    Problem is though that CSAP (our State's required test) still has a requirement of "Showing your Work", I believe. So I guess he does need to verbalize the answer (and write it) or perceive it (so it can be draw in picture form) or ?? I mean a picture is worth a thousand words. What is he suppose to put on paper if it is a requirement in our CSAP? Pictures I guess... Better than a thousand words!

  3. Heather, I had a terrific Algebra teacher who taught me something important: rather than thinking of it as "showing" my work, it's "double-checking" my work. I was so good at it that I wanted to do my problems very quickly and move on. It was boring. But this led to careless errors. When forced to slow down, I was able to achieve 100% accuracy every time. I still jumped to the answer, but then I went back and redid the problem a different way. As a consequence, rather than just making an A in the class, I made a 110. Kind of a silly difference then, but it helped me when I (much later) became a lawyer, because sometimes you need to be perfect. That's when it helps to build in some redundancy.