Home > Meeting Topics and Material, News Stories > QUEST Discussion Topics and News, Nov 30th

QUEST Discussion Topics and News, Nov 30th

QUEST 30 Nov 2012

QUEST Discussion Topics and News
30 Nov 2012:

We want to continue our discussion – QUEST for Autonomy. Specifically with respect to our conjecture – ‘to be autonomous requires the ability to respond to the unexpected query – and that requires ‘consciousness’. We’ve often referenced as a QUEST driver problem the data overload issue – one way to deal with this deluge of data is to select a small fraction of it and to process this reduced input in real-time, while the non-selected portion of the input is processed at a reduced bandwidth – *** but even this statement is misleading – as the conscious pathway appears to have far less bandwidth as a through channel than the unconscious pathways – so lets be careful how we state the goal – it is more accurate to suggest that the goal is to select a small fraction of the sensed data for a different type of processing – conscious processing versus more intense real-time processing *** this would require an autonomous ‘attention’ mechanism. This week we want to concentrate on the relationship between attention and consciousness – and we want to conclude the discussion with a focus on defining an experiment to be conducted in the Analyst test Bed on the topic – my going in position is we want it tied to ‘site monitoring’.

This is from the scholarpedia attention and consciousness: http://www.scholarpedia.org/article/Attention_and_consciousness

Table 1. A four-fold classification of percepts and behaviors depending on whether or not top-down attention is necessary and whether or not these percepts and behaviors necessarily give rise to phenomenal consciousness. Different percepts and behaviors are grouped together according to these two, psychophysically defined, criteria.
May not give rise to consciousness Gives rise to consciousness
Top-down attention is not required Formation of afterimages
Rapid vision (< 120 msec)
Accommodation reflex
Pupillary reflex
Zombie behaviors
Iconic memory
Animal & gender detection in dual-tasks
Partial reportability
Top-down attention is required Priming
Processing of objects
Visual search
Working memory
Detection and discrimination of unexpected & unfamiliar stimuli
Full reportability

Processing that requires top-down attention and results in consciousness
More than a century of research efforts have quantified the ample benefits that accrue to attended and consciously perceived events. For example Mack and Rock (Mack and Rock, 1998) compellingly demonstrated that subjects must attend to become conscious of novel or unexpected stimuli. These occupy the lower right quadrant of the attention x consciousness design matrix (Table 1). *** UNEXPECTED detection / discrimination requires attention AND consciousness * completely agrees with our recent position that to handle the unexpected query requires consciousness **

Last week we attempted to carve out a definition / position on ‘unexpected query’.

1.) Specifically defining an unexpected Query – to be autonomous implies being able to perform in dynamic missions in dynamic environments and do so in a manner acceptable to the team (other machine or human agents that are working towards and acceptable response to the current stimuli) – what we’ve termed as wingman solutions – Wingman solutions (solutions that a human can align with – interpredictability) require cognitive flexibility. Cognitive flexibility in the face of novelty requires consciousness! (QUEST definition of consciousness – or properties of it are listed below)
a. Cognitive flexibility is our term to capture the need to be able to respond appropriately to an “unexpected query”. Unexpected Query can be defined along the axes used by Dr. Stone in his Autonomy discussion – the axes of fidelity of reality known to the agent and the applicability of the stored models of the agent to react appropriately to the situation.
b. Cognitive flexibility also captures the idea that it may be necessary to consider alternative responses to the same stimuli. It may require:
i. Must explicitly represent / compare alternatives in context for world model used for perception (also for recollection and for imagination – point being there is a competition between plausible narratives that compete to become conscious) – Purely reactive systems do not use information with the same type of flexibility as deliberative systems, which can consider non-existent possibilities
ii. Must be able to deliberate with that representation / reason over previously non experienced phenomena (at a min ‘out of library’ – where characteristics defined in terms of prior experiences) – Define deliberate as – ability to represent and reason about nonexistent or unperceived phenomena (e.g., future possible actions or hidden objects) – Deliberative mechanisms include the ability to represent possibilities (e.g. possible actions, possible explanations for what is perceived) in some explicit form, enabling alternative possibilities (plausible narratives) to be compared and one selected
iii. Perceptual mechanisms produce information states between perception and action (intermediate states) usable in different ways in combination (plausible narratives) with different sorts of information, as required for human-like consciousness.

  1. No comments yet.
  1. No trackbacks yet.

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 )

Google+ photo

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

Twitter picture

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

Facebook photo

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


Connecting to %s

%d bloggers like this: