Brain Imaging Shows No Appreciable Neural Correlates of Memory Activity

There have been many brain scanning studies of how a brain looks when particular activities such as thinking or recall occur. Such studies will typically attempt to find some region of the brain that shows greater activity when some mental activity occurs. No matter how slight the evidence is that some particular region is being activated more strongly, that evidence will be reported and reported as a “neural correlate” of some activity.  But a question we should be asking is: do any such studies actually show appreciable evidence of any neural correlate of the activity under examination?

We should not be starting out by asking, "Which region of the brain changes most when a mental activity occurs?" The first and most fundamental thing to consider is: does there exist appreciable evidence of any correlation between brain states and higher mental activity? Similarly, it is a mistake to start asking, "Which person's face appears most commonly in the clouds?" It is much better to start with a simpler question such as "Is there appreciable evidence of any person's face appearing in the clouds?"

There are several types of memory activity that can be identified: 

(1) The acquisition of a new episodic memory through experience.

(2) The learning of a new physical skill by physical practice.

(3) The learning of new conceptual knowledge by school learning.

(4) Rote memorization, such as attempting to learn lists of words or names.

(5) The learning of a narrative by watching a play, TV show, or movie, or listening to a story being told.

(6) The recall of episodic memories a person has experienced.

(7) The recall of conceptual knowledge by someone answering a question or being asked to explain something. 

(8) Visual recognition, in which someone identifies some building, place or person. 

Although psychologists and neuroscientists often talk about "encoding," there is no understanding of any brain process by which knowledge is translated into synapse states or neural states. So when neuroscientists talk about "encoding" they are really just using a jargon word meaning "memory acquisition" or "learning." 

Let us look at whether there is any appreciable evidence of neural correlates for any of the eight activities listed above. 

Conceptual Learning or Memorization

• The study "Sustained Mnemonic Response in the Human Middle Frontal Gyrus during On-Line Storage of Spatial Memoranda" found no difference of more than about 1 part in 200 between different brain areas during a memorization task. 
• The study "Neural correlates of visual short-term memory for objects with material categories" found no difference of more than about 1 part in 200 between different brain areas during a memorization task. 
• The study "Neural correlates of encoding emotional memories: a review of functional neuroimaging evidence"  found no difference of more than about 1 part in 200 between different brain areas during a memorization task. 
• The study "Whole-brain functional correlates of memory formation in mesial temporal lobe epilepsy" found no difference of more than about 1 part in 200 between different brain areas during memory formation.
• The study "State-related and item-related neural correlates of successful memory encoding" found no difference of more than about 1 part in 200 between different brain areas during memory formation.
• The study "The neural correlates of recognition memory for complex visual stimuli in the Medial Temporal Lobe" found no difference of more than about 1 part in 250 between different brain areas during "memory encoding activity for faces and scenes," and about 1 part in 1000 for "memory retrieval activity for faces and scenes." 
• The paper "Neural correlates of multisensory perceptual learning" found no difference of more than about 1 part in 1000  between different brain areas

Memory Retrieval (Also Called Recollection)

• This brain scan study was entitled “Working Memory Retrieval: Contributions of the Left Prefrontal Cortex, the Left Posterior Parietal Cortex, and the Hippocampus.” Figure 4 and Figure 5 of the study shows that none of the memory retrievals produced more than a .3 percent signal change, so they all involved signal changes of merely about 1 part in 333 or smaller .
• In this study, brain scans were done during recognition activities, looking for signs of increased brain activity in the hippocampus, a region of the brain often described as some center of brain memory involvement. But the percent signal change is never more than .2 percent, that is, never more than 1 part in 500.
• The paper here is entitled, “Functional-anatomic correlates of remembering and knowing.” It shows a graph showing a percent signal change in the brain during memory retrieval that is no greater than .3 percent, less than 1 part in 300.
• The paper here is entitled “The neural correlates of specific versus general autobiographical memory construction and elaboration.” It shows various graphs showing a percent signal change in the brain during memory retrieval that is no greater than .07 percent, less than 1 part in 1000.
• The paper here is entitled “Neural correlates of true memory, false memory, and deception." It shows various graphs showing a percent signal change during memory retrieval that is no greater than .4 percent, 1 part in 250.
• This paper did a review of 12 other brain scanning studies pertaining to the neural correlates of recollection. Figure 3 of the paper shows an average signal change for different parts of the brain of only about .4 percent, 1 part in 250.
• This paper was entitled “Neural correlates of emotional memories: a review of evidence from brain imaging studies.” We learn from Figure 2 that none of the percent signal changes were greater than .4 percent,  1 part in 250.
• This study was entitled “Sex Differences in the Neural Correlates of Specific and General Autobiographical Memory.” Figure 2 shows that none of the differences in brain activity (for men or women) involved a percent signal change of more than .3 percent or 1 part in 333.
• A 2012 review study on "neural correlates of emotional memories" is one that we might expect to have a higher chance of showing a notable correlation, given the possibility of the emotions showing up as signal changes in the brain images. But the story reports no signal changes of greater than about 1 part in 1000 anywhere in the brain. 
• A brain scan study looked for neural correlates of "episodic retrieval success" during memory recall. The paper reports percent signal changes no greater than about 1 part in 500. 
• The study "Encoding Processes During Retrieval Tasks" found no difference of more than about 1 part in 300 between different brain states during episodic memory retrieval.
• The study "Neural activity associated with episodic memory for emotional context" found no difference of more than about 1 part in 200 between different brain states  during episodic memory retrieval.
• The paper "Parietal lobe contributions to episodic memory retrieval" found found no difference of more than about 1 part in 200 between different brain states during memory retrieval.
• The paper "Common and Unique Neural Activations in Autobiographical, Episodic, and Semantic Retrieval" found no difference of more than about 1 part in 200 between different brain states during memory retrieval.
• The paper "Functional-anatomic correlates of remembering and knowing" found no difference of more than about 1 part in 300 between different brain areas during memory retrieval.
• The paper "The role of the right prefrontal cortex in the retrieval of weak representations" found no difference of more than about 1 part in 500 between different brain areas during memory retrieval.

  Recognition Memory

• The year 2000 study "Dissociating State and Item Components
  of Recognition Memory Using fMRI" found no difference in brain signals of more than 1 part in 100, with almost all of the charted differences being only about 1 part in 500. 
• The study "Remembrance of Odors Past: Human Olfactory Cortex in Cross-Modal Recognition Memory" found no difference in brain signals of more than 1 part in 200.
• The study "Neural correlates of auditory recognition under full and divided attention in younger and older adults" found no difference in brain signals of more than 1 part in 500.
• The study "Neural Correlates of True Memory, False Memory, and Deception" asked people to make a judgment of whether they recognized words, some of which they had been asked to study. The study found no difference in brain signals of more than about 1 part in 300.
• The study "The Neural Correlates of Recollection: Hippocampal Activation Declines as Episodic Memory Fades" was one in which "participants performed a recognition task at both a short (10-min) and long (1-week) study-test delay." The study found no difference in brain signals of more than about 1 part in 300.
• The study "The neural correlates of everyday recognition memory" found no difference in brain signals of more than about 1 part in 500.
• The study "Neural correlates of audio‐visual object recognition: Effects of implicit spatial congruency" was one in which participants attempted a recognition task. The study found no difference in brain signals of more than about 1 part in 200.

We can summarize such results as follows: brains do not look any different and do not seem to act any different when a person is forming a new memory or recalling something previously learned or recognizing something previously encountered. Differences of  merely 1 part in 200 can be best explained as random fluctuations, the type of tiny blips that occur all the time in bodily things such as heart rate and breathing rate.  Such data is consistent with the idea that your brain is not the storage place of your memories, and that the formation and retrieval of memories is not a brain process. Also consistent with such an idea is the fact that no one has ever discovered a memory by examining brain tissue. No one has ever learned anything about a person's knowledge by examining the brain of a dead person. There is also no robust evidence for the storage of memories in animal brains. Claims to have detected memory storage spots in animal brains are junk science claims that do not hold up to critical scrutiny.  Typically an examination of the study group sizes used in any such study will show a failure to use study group sizes adequate to produce robust evidence. Neuroscientists lack any credible theory of how human episodic and conceptual knowledge could be translated into brain states or synapse states. What we know about synapses and the dendritic spines they are attached to (such as the less-than-monthly lifetimes of the proteins in such things, and their constant random remodeling) conflicts dramatically with claims that synapses could be a storage place of memories that can last for decades. 

Within the neuroscientist belief community that resembles a tribe or church, there is a pathological tradition under which signal variations of merely about 1 part in 200 are regarded as evidence for the brain being more actively engaged in some area. In no other field of biological study are variations so small regarded as good evidence. Let us imagine some scientist testing whether there is any truth to the common belief that your heart beats a little faster when you meet someone you are in love with. We can imagine a scientist hooking up heart rate monitors to young men, and analyzing the moments when young men met for a dinner date the female friends they were in love with. Now suppose the scientist found that the heart rate of such men only increased by 1 part in 200 at such meeting times (by an average of only about one third of a beat per minute).  How would this result be reported? It would be reported as a null result. The paper would claim that it had debunked the common idea that your heart beats faster when you meet your true love, and would say that the 1 part in 200 discovered was no significant evidence for such an effect. Only in the community of neuroscientists are 1 part in 200 signal change effects claimed as substantial evidence. In all other fields of biology, such a difference would be dismissed as negligible.

Let's imagine you are a neuroscientist who does some experimental brain scanning study looking for a neural correlate of some memory activity. You fail to find any appreciable evidence for such a thing, finding no difference of more than 1 part in 200 in brain activity. Now, you have a choice. You can either honestly write up your paper as a null result, using a title such as "Failure to find a neural correlate of recollection." But you know that in your neuroscientist community the habit of researchers is to report 1 part in 200 variations as positive results. And you know that science journals have a very big  publication bias, which is a strong tendency to prefer publishing papers reporting a positive result.  So do you do the honest thing decreasing your chance of paper publication (one that will irritate your colleagues by defying their customary behavior), or do you "go with the herd" and report your result as a "neural correlate"? Given the "publish or perish" culture in academia (in which the number of papers you publish and the number of citations they get is regarded as all-important), you may feel irrestible pressure to just follow the dysfunctional convention, and report your negligible correlation finding as a "neural correlate."  

You can get an idea of general conventions about correlation interpretation by doing a Google search for "guidlines for correlation interpretation." This will produce various papers like the one here, which give us interpretation guidelines such as this:

Size of correlation	Interpretation of correlation
.90 to 1.00	Very high correlation
.70 to .90	High correlation
.50 to .70	Moderate correlation
.30 to .50	Low correlation
.00 to .30	Negligible correlation

Clearly, following guidelines such as these, a percent signal change of only 1 part in 200 should be interpreted as a negligible correlation. Neuroscientists speak dishonestly when they try to pass off negligible results as being neural correlates of some kind of mind activity. 

The ability of neuroscientists to find correlation false alarms is illustrated in a 2021 paper entitled "Neurons in the mouse brain

correlate with cryptocurrency price: a cautionary tale."  The paper tells us this, referring to financial instruments mice cannot possibly know anything about:

"Out of ~40.000 recorded single neurons, ~70% showed a significant correlation with Bitcoin or Ethereum prices. Even when using the conservative Bonferroni correction for multiple comparisons, ~35% of neurons showed a significant correlation, which is well above the expected false positive rate of 5%."

After reading such a paper, you may realize how the "1 part in 200" signal changes typically reported in neural correlate studies are no robust evidence that brains are worker harder when someone learns or remembers anything.