Connectionists: Brains do not code time

[Juyang Weng]

I saw that Nature publisehd an Article by Albert Tsao et al. titled 
“Integrating time from experience in the lateral entorhinal cortex”.  
The authors are from Stanford University, (USA), NTNU of Norway, and 
Johns Hopkins University (USA).   Nature has published many such papers 
that show incremental advances in biology.

The abstract of the Article reads “in freely foraging rats that temporal 
information is robustly encoded across time scales from seconds to hours 
within the overall population state of the lateral entorhinal cortex.”

These claims by the authors are definitely wrong, because of authors’ 
great lack of overwarching understanding about how brains work and the 
severe limitations of their experiments.  The authors have designed very 
restricted task settings (in a box of four walls with two continuous 
sections of colors, black and white; the top section wall’s center has 
one color and the ramaining walls have the other color) and forced 
repeated experience (12 trials) onto the animal (rats)’s brains.  
Therefore, they greatly distorted what the multilemodal areas such as 
hippocampus do.   This is a severe disgrace to brain intelligence, 
contrary to what authors intended to do.  This Article is not an 
isolated case in terms of a narrow-minded approach to studying 
hippocampus or other brain areas.  Currently prevailing in biology, 
physiology, and medicine, this narrow-minded approach to studying brains 
is not going to lead us anywhere interesting.

In particular, none of the brains areas in general,  hippocampal areas 
in particular, code time at all!  They only deal with sensorimotor 
events like an emergent Turing machine.   The brains must deal with 
time-warping, like an automata, not directly encoding time!   However, I 
am afraid none of the authors are familiar with emergent Turing machines.

June 13, 2017, we submitted two joint Letters to Nature: Juyang Weng, “A 
Model for Auto-Programming for General Purposes”, and Juyang Weng, Zejia 
Zheng, Juan L. Castro-Garcia, Xiang Wu, Qian Guo, and Xiaofeng Wu, “Some 
Experiments towards Auto-Programming for General Purposes”.  Nature 
rejected both Letters June 20, 2017.

The email from Nature’s Senior Editor Leonie Mueck stated:

“I regret that we are unable to publish it in Nature.  As you may know, 
we decline a substantial proportion of manuscripts without sending them 
to referees, so that they may be sent elsewhere without delay. In such 
cases, even if referees were to certify the manuscript as technically 
correct, we do not believe that it represents a development of 
sufficient scientific impact to warrant publication in Nature. These 
editorial judgements are based on such considerations as the degree of 
advance provided, the breadth of potential interest to researchers and 
timeliness. ”

“In this case, we do not feel that your paper has matched our criteria 
for further consideration. We therefore feel that the paper would find a 
more suitable outlet in another journal.”

There are no paper specific comments in the review.  In particular, none 
of the comments in the email address the major contributions of the 
papers at all.

The cover letter of our first Letter to Nature provided a 100 word or 
less summary indicating on scientific grounds why the paper should be 
considered for a wide-ranging journal like Nature instead of a more 
narrowly focused journal:

“The results here have wide-front implications to computer science and 
engineering.   The Turing Machine theory has yet to be tightly 
associated with brain-like network computation.   While a Universal 
Turing Machine is considered for general purposes, a human must program 
it for each purpose.   It was not known that a non-human machine can 
automatically program for general purposes, only a human adult can. In 
this sense, the new machine is a general-purpose self-programming Turing 
Machine, different from
the (irrational number capability, or from rational numbers to real 
numbers) sense of Hava Siegelman 1995 published in  Science.”

“… those who like to duplicate and expand the simulation results on 
machines must read this theoretical letter because it is impossible to 
duplicate the results without understanding the theory.   The same seems 
to be true if one likes to biologically verify the theory here. “

After much writing improvements, January 2, 2018 we submitted two joint 
reports to the Science Magazine but the worse events have happened: Not 
only did Science not provide any review comments at all, after the 
rejection by Science Magazine our web site was cyber-attacked probably 
because the site criticized that Science Magazine violated COPE rules!  
Please watch the YouTube video: “BMTalk 3D Episode 2: Science Magazine 
Rejected GENISAMA Super Turing Machines”, https://youtu.be/Qf8qjgBMasc

It is unfornate that human scientists are too narrow-minded for the 
subjects that they study, blocked by cross-disciplinary journals like 
Science and Nature.  Those who do research on biological brains do not 
really care how brains compute overall; those who do artificial 
intelligence do not really care how brains work.   Much taxpayers’ money 
have been spent wastefully on many such narrow-minded and shallow 
projects, while our brain-model GENISAMA Turing Machines and 
experimental works have already explained, shown, and demosntrated how 
brains work.  Yet, the Science and Nature journals presistently block 
much needed cross-disciplinaery scientific communications.

Again, brains do not represent time, but sensorimotor events instead.   
The human race took over two decades to wake up for the first deep 
learning network (Cresceptron 1992, 1993) with which we started 
general-purpose visual learning from 3D worlds.  How many more decades 
does the human race needs before it wakes up and accepts this much  
greater breakthrough in Natural Intelligence and Artificial 
Intelligence–GENISAMA Super Turing Machines?  What should I do?  What 
should scientists do?

-- 
--
Juyang (John) Weng
------------------- Work ---------------------  ---- Technology Transfer ----
Professor                                       Founder
Department of Computer Science and Engineering  GENISAMA LLC
MSU Cognitive Science Program                   Okemos, MI 48864 USA
and MSU Neuroscience Program                    Tel: 517-980-6270
428 S Shaw Ln Rm 3115                           Web: genisama.com
Michigan State University                       --------- Outreach ----------
East Lansing, MI 48824 USA                      Founder
Tel: 517-353-4388                               Brain-Mind Institute
Fax: 517-432-1061                               Web: brain-mind-institute.org
Email: weng at cse.msu.edu                         Brain-Mind Magazine
Web: http://www.cse.msu.edu/~weng/              Web: brain-mind-magazine.org
----------------------------------------------  -----------------------------