Mind-modelling: Marking and Feedback

Our last two posts have been focused on mind-modelling (Stockwell 2009), and in this post we’re going to think about how it can offer insights into practices and challenges surrounding marking and feedback. Mind-modelling – the process of constructing a working model for what we think is going on in another person, or character’s, mind – is a practice we engage in when we read literature, regardless of genre, all the time. In fact, we could write an entirely separate post on this (and if you’d like us to, let us know!) because, unlike in life, in fiction we can actually be offered a window into character’s thoughts, feelings and mental processes. Authors often very deliberately manipulate the level of access readers have to particular character minds: Severus Snape in Harry Potter and Mr Darcy in Pride and Prejudice are two great examples where access to their true thoughts and feelings is very carefully withheld, and it’s no accident that both are typically inaccurately mind-modelled by both readers and other characters for most of these stories as a result.

It’s easy for us to forget that in real life, our sense of what another person is thinking is based on our best guess: whenever we interact with others we are running a dynamic mind-model based on all sorts of verbal and non-verbal cues, and often moderating our own behaviour according to what we think is going on in the other person’s head.

A mind-model which does not actually represent a good match to another person’s actual mind is at the root of many arguments and misunderstandings. Imagine a friend passes you in the street, looks at you, and completely ignores you. Your mind-modelling might go into overdrive, constructing that they have feelings of anger or annoyance towards you. You might begin wracking your brain to think what you could have done to provoke this reaction. You might even come up with something and either decide that it’s unreasonable, and then act angrily towards them or ignore them yourself the next time you meet, or feel remorse and perhaps call or text them to apologise. All the time, your friend was deep in thought about the latest episode of The Handmaid’s Tale and had forgotten to put their contact lenses in that morning. Here, well-reasoned but completely inaccurate mind-modelling has led to unsuccessful communication, and it’s easy to see, if you did reciprocate angrily towards your friend on your next meeting, how a series of miscommunications can escalate and even put the friendship in jeopardy.

Marking and feedback possesses all the hazards of inaccurate mind-modelling between teachers and students that we constantly traverse in day-to-day interaction. But with marking and feedback the situation is actually often much more precarious, for several reasons:

Classrooms and schools are inherently sites of unequal encounter (Fairclough 2014)

The power imbalance between teacher and students is always likely to subtly affect the nature of interaction, especially in terms of what students are willing or not willing to say. There’s no magic bullet to fix this issue, but we think that simply being mindful of it can be really useful. In terms of marking and feedback, This means that students are predisposed not to ask whether their mind-model for their teacher is accurate or not. This is like if, in the scenario above, you never ask your friend if or why they were ignoring you. In day-to-day interaction this is a key way in which disputes can escalate and relationships can break down, so it seem uncontroversial to suggest that there’s a risk of this happening in the classroom too.

Marking as Attenuated Interaction

Marking a student’s work involves several stages of mind-modelling. This is a highly attenuated process: each stage offers another opportunity for misinterpretation. In essence, the stages, through the lens of mind-modelling often look like this:

  1. Teacher tries to mind-model what the student will think when reading comments, viewing underlines, ticks, crosses and so on.
  2. Student, with little other than the teacher’s marking to go on, tries to mind-model what the teacher was thinking as they marked the piece of work and interprets the feedback accordingly.

If there’s a mismatch between mind-models there’s very little opportunity for this to be picked up.

What about dialogic approaches?

The idea of dialogic feedback, where students respond to teachers’ comments and perhaps then teachers even respond to those and so on, is relatively sound in principle. However, as when you’re writing your colleague a long email and you decide to just give up and call them instead, it’s difficult to see how a written dialogic approach can be as good as an actual conversation. Added to this a conversation is likely to take less time and offers more opportunities to make sure your understanding of what a student is struggling with is accurate.

Recently Pinkett wrote a great account of strategies he has employed in his own classroom, suggesting replacing written feedback with one-to-one sessions with students talking through their work with them:

twice weekly, midway through a lesson, once I’ve set the students off on an extended writing task (18 minutes minimum), I haul a desk to the front of the classroom and sit myself down on a chair under the whiteboard, facing the class. […] One by one, students ‘come up’ and talk me through some of the work in their exercise books. They turn the pages and read sections of their efforts to me.’

(TES, Pinkett 2016)

From a mind-modelling perspective, Pinkett’s approach makes a huge amount of sense. Whilst marking provides a written record that feedback has been given, it seems a much less reliable way of conveying accurate feedback to students than having a conversation with them if that conversation offers good opportunities for accurate mind-modelling on both sides. If your school insists on a written record, could you adopt Pinkett’s approach and then ask students to write up the key points of your discussion when they return to their desk?

Verbal feedback also offers teacher and students opportunities for queries and clarification not typically offered by written feedback. Here additional opportunities for catching misinterpretations by recalibrating mind-models for one another are introduced into the teacher-student interaction.


A key issue with providing written marking and feedback is, as every teacher is well aware, the huge amount of time it can take. A particular issue with this in terms of mind-modelling, however, is what we can tend to cut out of the feedback in order to make the job a little quicker. Three things that are naturally susceptible to the chop are:

  1. Hedging: ‘There are some slight issues with your style’ can become ‘Issues with style’
  2. Complete syntax/fuller explanations: ‘This paragraph isn’t adding anything to your argument, you could cut it out’ can become ‘Cut’ or even just a line through the paragraph.
  3. Positive comments: Whilst most teachers agree that it’s nice to have positive comments to balance out critiques, faced with 50 essays on a weekend day and reflecting on what is really going to help students improve, positive comments can quickly seem like a time cost that there simply isn’t room for.

One issue with this natural attrition of these stylistic features of a teacher’s feedback is that if a student is using the marking that remains to construct a mind-model of what the teacher thought of their essay the loss of these features can result in inaccurate modelling that teacher is rude, teacher doesn’t like me, teacher thinks this essay is rubbish, teacher thinks I am rubbish. When this is compounded with the power imbalance meaning that students are less likely to actually ask their teacher if these modelling are accurate, this can mean that students become resistant to taking any feedback on board at all. We are certainly not suggesting that teachers should therefore double their marking time by keeping these features in but we think that being aware of these dangers is really valuable.

Some strategies that might be helpful
  • Could you help students to understand why these features are absent through meta-narration: ‘you might look at some of the comments on your work and think that they seem a bit rude, so I just want to take a moment to discuss this…’. This is a good potential strategy for two reasons. First, it takes much less time to reinsert all of the hedging, positivity and full explanations verbally that get lost when they’re written down. Second, it offers students more information with which to construct their mind-model for you, which can head off the inaccurate modelling before it occurs.
  • There seems to have been a recent shift towards whole class feedback – from a mind-modelling point of view this makes a lot of sense because it decreases the level of threat posed by criticisms to individual students.
  • Verbal interactions about work create more opportunities for both teacher and student(s) to establish an accurate mind-model for one another and, crucially, for inaccurate modelling to be identified and redressed.
  • Could you experiment with getting some students to do what researchers call a think-aloud protocol with some of your marking and feedback. This would mean getting students to silently read through their essay with your comments and asking them to say what they’re thinking as they read about what they think you’re asking for and what you think of the piece of work. This might be an interesting way to try to see how your students are mind-modelling you and could even be used as a foundation to engage your classes in discussion about marking and feedback, it’s advantages, pitfalls and how everyone could make it work more effectively.

We hope you’ve found this cognitive perspective on marking helpful, let us know your thoughts in the comments or on Twitter at @studyingfiction


Fairclough, N. (2014) Language and Power. Second Edition. London: Routledge.

Pinkett, M. (2016) ‘The teacher who gave up marking and thinks that you can too’, Times Educational Supplement. Available online here: https://www.tes.com/news/school-news/breaking-views/teacher-gave-up-marking-believes-you-can-too

Stockwell, P. (2009) Texture: The Cognitive Aesthetics of Reading. Edinburgh: Edinburgh University Press.







One thought on “Mind-modelling: Marking and Feedback

  1. Another good post, Jess. I’ve many misgivings about written feedback about written feedback from my time as a TA that I wouldn’t commit to print! Face-to-face interaction is always better in my experience but it’s the damn time. Eight weeks of office hours spent twiddling thumbs before finding stuff to get on with, then with summative coursework looming, a queue starts forming and you’re conducting additional consultation hours after classes and in lunch breaks… Usually courses had formative cw too, but then if it’s formative and other work on other courses needs doing, you’d get a handful of those. And you need a certain amount of weeks on a module before confident about writing even a formative piece. I really like the strategies though and think more seminars used for workshopping written pieces and feedback (and looking at one another’s work) would be more useful than ‘what did you make of those six articles on the readling list?’ would be time better spent.

    I also enjoyed ‘All the time, your friend was deep in thought about the latest episode of The Handmaid’s Tale and had forgotten to put their contact lenses in that morning.’ More of that please and perhaps I should get round to watching that series.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s