AdamWTeach
@adamwteach.bsky.social
Science Teacher of Autistic students, ex tech engineering exec, into latest cognitive science - Active inference predictive coding and how the mind really stores and retrieves memory. Geeky about data and assessment. Blog: predictablycorrect.com
Please find my response ... i'm a bit prone to verbosity so had to do a blog!! #EruditePedagogy
open.substack.com/pub/predicta...
open.substack.com/pub/predicta...
Load to Prediction - Variation Theory
In response to Christian Moore-Anderson's question on how Predictive Processing / Active Inference theory applies to aspects of Variation Theory
open.substack.com
June 29, 2025 at 4:07 PM
Please find my response ... i'm a bit prone to verbosity so had to do a blog!! #EruditePedagogy
open.substack.com/pub/predicta...
open.substack.com/pub/predicta...
..... thinking.... I am taking the time to read your book ( Difference Maker) so I can better address your excellent question, thinking of making a concrete example, e.g. using your speaking Russian and then discovering foreign languages.... I struggle with brevity, so need to publish as a blog.
June 27, 2025 at 10:29 PM
..... thinking.... I am taking the time to read your book ( Difference Maker) so I can better address your excellent question, thinking of making a concrete example, e.g. using your speaking Russian and then discovering foreign languages.... I struggle with brevity, so need to publish as a blog.
However, it links to PP in terms of the brain evolving as a prediction machine. The "forgetting" mechanism, the design of the biochemical cascades and messaging, the effect of dopamine, pyramidal neuron structure are all the evolved structure that is able to infer salience from predict. errors *6
June 27, 2025 at 10:00 PM
However, it links to PP in terms of the brain evolving as a prediction machine. The "forgetting" mechanism, the design of the biochemical cascades and messaging, the effect of dopamine, pyramidal neuron structure are all the evolved structure that is able to infer salience from predict. errors *6
I agree that PP/ActInf does not feel to give clear explanatory support to why retrieval practice and spacing are effective. I believe that comes from understanding the neurochemical processes around how engrams- networks of neurons -synapses, receptors, dendritic spines etc. all actually work. *5
June 27, 2025 at 9:55 PM
I agree that PP/ActInf does not feel to give clear explanatory support to why retrieval practice and spacing are effective. I believe that comes from understanding the neurochemical processes around how engrams- networks of neurons -synapses, receptors, dendritic spines etc. all actually work. *5
In addition the fact that desirable difficulty and managing limits to learning load can fit together with the same neuro science backed mechanism fits an occam's razor principle for me *4
June 27, 2025 at 9:50 PM
In addition the fact that desirable difficulty and managing limits to learning load can fit together with the same neuro science backed mechanism fits an occam's razor principle for me *4
re. Desirable difficulties, my point is that "difficulty" is subject to interpretation, and lots of work has gone into trying to pin it down in terms of element interactivity etc. but I find the PP model of large enough surprise to trigger model updating compelling and backed with neuroscience. *3
June 27, 2025 at 9:49 PM
re. Desirable difficulties, my point is that "difficulty" is subject to interpretation, and lots of work has gone into trying to pin it down in terms of element interactivity etc. but I find the PP model of large enough surprise to trigger model updating compelling and backed with neuroscience. *3
... If PP/Act inf is used as the base cog architect, rather than lim. working mem. the level of prior knoweldge as the major effect is a direct outcome. in fact complexity of task design was also another factor, but this is displaced by student familiarity (predicticability) of the task design *2
June 27, 2025 at 9:43 PM
... If PP/Act inf is used as the base cog architect, rather than lim. working mem. the level of prior knoweldge as the major effect is a direct outcome. in fact complexity of task design was also another factor, but this is displaced by student familiarity (predicticability) of the task design *2
The point here was the Expert reversal effect was a sticking plaster to explain why "cognitive Load" didn't increase with element interactivity. In fact findings were that prior knowledge was the main factor in the self reported feeling of "load" - *1
June 27, 2025 at 9:37 PM
The point here was the Expert reversal effect was a sticking plaster to explain why "cognitive Load" didn't increase with element interactivity. In fact findings were that prior knowledge was the main factor in the self reported feeling of "load" - *1
#EruditePedagogy .. let's start... @cmooreanderson.bsky.social you felt there were some areas of variation theory that Predictive processing /active inference theory don't address/explain/support... What are your thoughts here? I like to understand and develop that thought.
June 26, 2025 at 4:33 PM
#EruditePedagogy .. let's start... @cmooreanderson.bsky.social you felt there were some areas of variation theory that Predictive processing /active inference theory don't address/explain/support... What are your thoughts here? I like to understand and develop that thought.
@cmooreanderson.bsky.social I have started a Note on substack as a place to develop this discussion on variation theory - I would like to understand and discuss which parts of variation theory you feel that PP/ActInf don't explain well. You have more extensive knowledge on Var theory than I do!
June 25, 2025 at 8:11 PM
@cmooreanderson.bsky.social I have started a Note on substack as a place to develop this discussion on variation theory - I would like to understand and discuss which parts of variation theory you feel that PP/ActInf don't explain well. You have more extensive knowledge on Var theory than I do!
I li think the combo of pp/act inf and the neurobiochemisty will help but not be the whole picture of the mechanism by which the brain updates it's generative model.. how does it infer statistical patterns and generalities? This may well link to and be informed by why variation theory works
June 25, 2025 at 6:51 PM
I li think the combo of pp/act inf and the neurobiochemisty will help but not be the whole picture of the mechanism by which the brain updates it's generative model.. how does it infer statistical patterns and generalities? This may well link to and be informed by why variation theory works
When I say guts of it I mean the system/ fundamental mechanisms, that the effects, instructional theories and pedalogical strategies sit upon. But I think we also need to go the layer lower into the biochemical processes of synapses and neurons to understand more on how learning happens.
June 25, 2025 at 6:48 PM
When I say guts of it I mean the system/ fundamental mechanisms, that the effects, instructional theories and pedalogical strategies sit upon. But I think we also need to go the layer lower into the biochemical processes of synapses and neurons to understand more on how learning happens.
I've skimmed the difference maker, and agree with the fact that there have been many of the ideas and philosophies that have come before that have likely been some cornerstones, steps towards, supporting acts to PP/actinF. It's my feeling that PP/ActInf is getting to the guts of it all.
June 25, 2025 at 6:45 PM
I've skimmed the difference maker, and agree with the fact that there have been many of the ideas and philosophies that have come before that have likely been some cornerstones, steps towards, supporting acts to PP/actinF. It's my feeling that PP/ActInf is getting to the guts of it all.
In fact Christian you comments on variation theory being an instructional theory vs. prediction processing being a theory of cognition helped shape this blog.
June 25, 2025 at 6:41 PM
In fact Christian you comments on variation theory being an instructional theory vs. prediction processing being a theory of cognition helped shape this blog.
Christian, many thanks for your comments, I had the feeling that this is where your comments thus far were coming from.. I embrace them and have already started a blog in responding... I.e. furthering the socratic devabate
June 25, 2025 at 6:40 PM
Christian, many thanks for your comments, I had the feeling that this is where your comments thus far were coming from.. I embrace them and have already started a blog in responding... I.e. furthering the socratic devabate
Challenge accepted..
June 14, 2025 at 10:34 PM
Challenge accepted..
I'm prepping a new substack on it now. Just trying to balance nerdiness to accessibility!
June 14, 2025 at 9:28 PM
I'm prepping a new substack on it now. Just trying to balance nerdiness to accessibility!
On the point of engrams and neuronal firing I have done a deep research dive in how our brains do store /update it's information (we'll update it's generative model). Is fascinating, but also informs spacing/retrieval and has some unique implications for teaching I think
June 14, 2025 at 9:26 PM
On the point of engrams and neuronal firing I have done a deep research dive in how our brains do store /update it's information (we'll update it's generative model). Is fascinating, but also informs spacing/retrieval and has some unique implications for teaching I think
But I do think that it's incumbent on us, and especially on those thought leaders to take note of and understand if there are new and improved theories of how our brains work, and how best to help them learn to best work in the future world our students are going into.
June 14, 2025 at 9:16 PM
But I do think that it's incumbent on us, and especially on those thought leaders to take note of and understand if there are new and improved theories of how our brains work, and how best to help them learn to best work in the future world our students are going into.
In teaching I am sure the likes of @adamboxereducation.bsky.social tom Sherrington @teacherhead.bsky.social Craig Barton, Ollie Lovell, Sarah Cottinghatt etc who are in lots of schools would say the biggest bang for the buck is in basic behaviour / good practice.
June 14, 2025 at 9:13 PM
In teaching I am sure the likes of @adamboxereducation.bsky.social tom Sherrington @teacherhead.bsky.social Craig Barton, Ollie Lovell, Sarah Cottinghatt etc who are in lots of schools would say the biggest bang for the buck is in basic behaviour / good practice.
I achieved some the best yield/quality improvements by doing the simple stuff, but over time the theoretical stuff gave some step changes in capability. I think you need both.
June 14, 2025 at 9:09 PM
I achieved some the best yield/quality improvements by doing the simple stuff, but over time the theoretical stuff gave some step changes in capability. I think you need both.