Jump to ContentJump to Main Navigation
Unlocking the BrainVolume 1: Coding$

Georg Northoff

Print publication date: 2013

Print ISBN-13: 9780199826988

Published to Oxford Scholarship Online: April 2014

DOI: 10.1093/acprof:oso/9780199826988.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 27 May 2017

Appendix 1 Neuroempirical Remark: Resting-State Activity versus Stimulus-Induced Activity—Continuity Hypothesis (p.299)

Appendix 1 Neuroempirical Remark: Resting-State Activity versus Stimulus-Induced Activity—Continuity Hypothesis (p.299)

Source:
Unlocking the Brain
Publisher:
Oxford University Press

Summary

I discussed the neuronal and biochemical mechanisms underlying the transition from resting-state to stimulus-induced activity in Part IV. This showed that the very same neuronal and biochemical mechanisms operating in the resting state are also at work during the transition from resting-state to stimulus-induced activity; that is, rest–stimulus interaction. This leads me on a theoretical level to describe what I call the “continuity hypothesis.” The continuity hypothesis is a hypothesis about the relationship between resting-state and stimulus-induced activity. I postulate neuronal continuity between resting state and stimulus-induced activity, with both being continuous and discontinuous in their neural activities at the same time. The co-occurrence of both neuronal continuum and discontinuum between resting-state and stimulus-induced activity is supposed to be made possible by difference-based coding, which in turn is regarded as essential for enabling and predisposing consciousness. The “continuity hypothesis” bridges the often-presupposed divide between intrinsic and extrinsic characterizations of the brain as put forward in the Introduction: both views are no longer considered opposite and contradictory, but rather, two extremes of an underlying continuum of neural activity amounting to an intrinsic-extrinsic view of the brain.

Key Concepts and Topics Covered

Resting-state activity, stimulus-induced activity, neuronal continuum and discontinuum, difference-based versus stimulus-based coding, degree versus origin, neural predisposition versus neural correlate, consciousness

Neuroempirical Remark IA: Difference-Based Coding as Unifying Code between Resting State and Stimulus-induced Activity

The brain is often considered an either purely extrinsic or intrinsic organ (see also Introduction). In the case of an extrinsic characterization, the brain is supposed to be characterized by stimulus-induced activity as it is related exclusively to the stimulus itself. In contrast, an intrinsic view proposes intrinsic activity in the brain, that is, resting-state activity, with its function remaining unclear.

How do intrinsic and extrinsic views of the brain stand in relation to each other? Are they mutually exclusive and thus incompatible? Or, rather, are they compatible and complementary? This is the topic here, and it will be discussed in the framework of what I describe as the “continuity hypothesis.” The “continuity” hypothesis implies complementarity and compatibility between resting-state activity and stimulus-induced activity rather than incompatibility and opposition.

Let me now sketch the “continuity hypothesis” in more detail. I argued that stimuli of different origins, intero- and exteroceptive and neural, are encoded into neural activity in the same way; namely, in terms of spatial and temporal differences. This implies that both resting-state activity (see Chapters 4–6) and (p.300) stimulus-induced activity (see Chapters 10–12) are based on difference-based coding rather than stimulus-based coding.

That also entails that the various interactions between different stimuli, that is, stimulus–stimulus interaction (see Chapter 10); between stimuli and resting state, that is rest–stimulus interaction (see Chapter 11); and the interactions within the resting state itself, that is, rest–rest interaction (see Chapter 4 and 5), were encoded in terms of one and the same code; namely, difference-based coding.

What does this tell us about the role of difference-based coding in the brain? I postulate difference-based coding to be the general coding strategy of the brain’s neural activity during its different kinds of interactions: that is, rest–rest, rest–stimulus, stimulus–stimulus, and stimulus–rest. Since all neural activity is supposed to arise from these interactions, difference-based coding must be considered the neural code of the brain.

This means that rest–rest, rest–stimulus, stimulus–stimulus, and stimulus–rest interactions are linked and united by one coding strategy—difference-based coding. I thus propose that the application of that code to both resting-state and stimulus-induced activity makes their direct interaction possible.

Neuroempirical Remark IB: “Continuity Hypothesis” about the Neuronal Relationship between Resting State and Stimulus-Induced Activity

How can we further specify the relationship between the brain’s resting-state activity and its stimulus-induced activity? Due to the fact that resting-state and stimulus-induced activity operate on the basis of one and the same neural code, there must be neuronal continuity between them: that is, between and across rest–rest, rest–stimulus, stimulus–stimulus, and stimulus–rest interaction.

Empirically, such assumption is, for example, supported by the observations showing the resting state’s strong low-frequency fluctuations and functional connectivity to be carried forth into stimulus-induced activity via rest–stimulus interaction (see Chapters 4, 5, and 11). Conversely, the data also show that the stimulus-induced activity resurfaces in the neural pattern of the subsequent resting-state activity, which is supported by the data on stimulus–rest interaction (see Chapter 11).

I consequently postulate what I describe as a “continuity hypothesis” between resting-state activity and stimulus-induced activity. The continuity hypothesis describes that neuronal activity is continuously carried back and forth between resting-state and stimulus-induced activity (via rest–stimulus and stimulus–rest interaction), resulting in neuronal continuity between both forms of neural activity.

The “continuity hypothesis” is a purely neuronal hypothesis that pertains to the neuronal features and characterization of resting state and stimulus-induced activity. In contrast to the neuronal states of the brain during resting state and stimulus-induced activity, the “continuity hypothesis” does not make any assumptions about the behavioral, psychological, and phenomenal states that are associated with the resting state and stimulus-induced activity. Hence, the neuronal continuity between resting state and stimulus-induced activity does not imply behavioral, psychological, or even phenomenal continuity.

This, however, does not mean that the continuity hypothesis does not carry important implications for behavioral, psychological, and phenomenal states. We will see in Volume II that the neuronal continuity between resting state and stimulus-induced activity is essential in making possible the association of a phenomenal state—consciousness—with the otherwise purely neuronal resting state or stimulus-induced activity (see especially Chapter 30).

Neuroempirical Remark IC: Neuronal Continuity between Resting State and Stimulus-Induced Activity

The “continuity hypothesis” must be further specified by two aspects, “neuronal continuum” and “neuronal discontinuum.” “Neuronal continuum” describes that neural activity in resting state and stimulus-induced activity show similar features, accounting for a neuronal continuum (p.301) between them. The concept of “neuronal discontinuum” refers to the differences between resting-state and stimulus-induced activity operating across and thus superseding their underlying neuronal continuum.

How can we now describe both neuronal continuum and neuronal discontinuum in further detail? Let me start with the neuronal continuum. The attentive reader may have noticed that I focused on the same kind of neuronal measures when discussing resting-state activity (Part II) and stimulus-induced activity (Part IV). In both cases, I described spatial and temporal measures of neural activity; more specifically, functional connectivity and low- and high-frequency fluctuations.

Functional connectivity and frequency fluctuations are central in constituting both resting-state and stimulus-induced activity; furthermore, they mediate their direct interaction, that is, rest–stimulus and stimulus–rest, via the principles of spatial and temporal coincidence. As such, functional connectivity and frequency fluctuations signify a neuronal continuum between both forms of neural activity.

However, the neuronal continuum went further. Due to the reliance on the same spatial and temporal measures, that is, frequency fluctuations and functional connectivity, both resting-state and stimulus-induced activity are supposed to be linked by a shared and common spatiotemporal structure (see Chapters 4 and 5). Such a spatiotemporal structure operates across and supersedes the purely biophysical-computational features of the brain’s space and time in that it is statistically based rather than being exclusively determined by the biophysical-computational features of the brain’s neurons (and regions; see Chapters 1, 2, 6, and 11 for details).

The statistically based spatiotemporal structure of the brain’s resting state is carried forth to stimulus-induced activity (via rest–stimulus interaction), which in turn impacts the subsequent resting-state activity (via stimulus–rest interaction; see Chapter 11). Such circular movement between resting-state activity and stimulus-induced activity allows for maintaining (and continuously rejuvenating and updating) the brain’s intrinsic spatiotemporal structure. One may consequently characterize the brain’s intrinsic spatiotemporal structure as the common final functional pathway of both resting-state and stimulus-induced activity. As such, the spatiotemporal structure can provide a neuronal continuum between the two forms of neural activity (see Fig. A1-1a).

Neuroempirical Remark ID: Neuronal Discontinuity between Resting State and Stimulus-Induced Activity

So far, I have focused on the similarities and thus the neuronal continuum between resting-state and stimulus-induced activity. This, however, should not incline us to brush over their considerable differences accounting for a neuronal discontinuum between the two forms of neural activity. This neuronal discontinuum shall now be further specified. We recall from Chapter 11 that rest–stimulus (and also stimulus–rest) interactions were characterized by non-linearity; non-linearity describes that the resulting stimulus-induced activity does not result from mere linear addition or superposition of both forms of neural activity. The resulting stimulus-induced activity is consequently different both spatially and temporally from the preceding resting-state activity. Hence, there is a neuronal discontinuum on the neuronal level between both forms of neural activity.

The neuronal discontinuum is also visible in the principle of inverse effectiveness (see chapters 10 and 11). In a nutshell, the principle of inverse effectiveness describes that a lower resting-state activity level may lead to stronger rest–stimulus interaction in the presence of a strong stimulus, compared to higher resting-state activity level in the presence of the same stimulus. While empirical support for this principle is mostly pending, it clearly signifies a neuronal discontinuum. A lower resting-state activity turns into a stronger stimulus-induced activity, thus making the latter more discontinuous from the former. In contrast, there is a lower degree of the neuronal discontinuum when the resting-state activity is higher. (p.302)

Appendix 1 Neuroempirical Remark: Resting-State Activity versus Stimulus-Induced Activity—Continuity HypothesisAppendix 1 Neuroempirical Remark: Resting-State Activity versus Stimulus-Induced Activity—Continuity Hypothesis

Figure A1-1 “Continuity hypothesis” between resting state and stimulus-induced activity. The figure depicts two central aspects of the continuity hypothesis, the conjunction of both neural continuum and neural discontinuum (a), and the central role of the resting state and its spatiotemporal structure as neural predisposition (b).

(a) The figure shows the neural continuum (upper part) and discontinuum (lower part) between resting-state activity and stimulus-induced activity. Both resting-state activity and stimulus-induced activity are based on the same spatiotemporal measures (like functional connectivity and low-frequency fluctuations) and apply the same neural code, difference-based coding and sparse coding (upper part). At the same time, though, there is also neural discontinuum (lower part) between resting-state activity and stimulus-induced activity in that there is non-linearity, inverse effectiveness, and variation in the degrees of difference-based coding and sparse coding.

(b) The figure depicts how the stimulus modifies the resting state’s spatiotemporal structure, as indicated in the upper part of the figure. However, the degree of modification the stimulus can possibly elicit is predisposed by the resting state itself and its spatiotemporal structure. This means that the resting state provides a neural predisposition, a “spatiotemporal window of opportunity, ” for the possible degree of subsequent stimulus-induced activity (lower part of the figure).

(p.303)

How can gather further support for the neuronal discontinuum between resting state and stimulus-induced activity? I also demonstrated that the degree of both difference-based coding and sparse coding changes during the encounter with the stimulus. Rest–stimulus interaction may go along with a shift in the balance between difference- and stimulus-based coding toward difference-based coding. There is thus a neuronal discontinuum with regard to the degrees of difference-based coding.

The same holds for the formatting. As tentatively suggested, the degree of sparse coding may increase during stimulus-induced activity when compared to resting-state activity, thereby leading to a neuronal discontinuum on a formatting level (see Chapter 12). Finally, the neuronal discontinuum also extends to the biochemical level, with GABA and glutamate showing different degrees of difference in the excitation-inhibition balance during resting state and stimulus-induced activity (see Chapters 2, 6, and 12).

Neuroempirical Remark IIA: Continuum of Neuronal Measures

How are the neuronal continuum and neuronal discontinuum related to each other? I postulate that the neuronal continuum provides the very basis upon which the neuronal discontinuum operates. More specifically, the degrees of the resting state’s functional connectivity and high-low frequency fluctuations are varied during subsequent stimulus-induced activity (see Chapter 11).

The same holds for the changes in the degrees of difference-based coding and sparse coding as well as for the degrees of the spatiotemporal structure. As detailed in Chapters 11 and 12, the stimulus may introduce a novel degree of discontinuity into these neuronal measures of the brain’s resting-state activity: by varying the resting state’s diverse spatial and temporal neuronal measures in their degree, the stimulus introduces a much higher degree of discontinuity compared to the dynamic changes in the resting-state activity itself.

This means that the neuronal discontinuum and thus the neuronal difference between resting-state activity and stimulus-induced activity is a matter of degree. This implies that there is no principal difference between resting-state activity and stimulus-induced activity. Instead of being principally (and qualitatively) different, the stimuli and their associated stimulus-induced activity operate across and supersede resting-state activity and modulate it quantitatively; the stimulus “uses” the resting state’s diverse neuronal measures as a starting point to modulate and vary them in their degrees.

Accordingly, the neuronal measures themselves, like functional connectivity and low frequency fluctuations, thus provide the neuronal continuum between resting state and stimulus-induced activity, while their degree signifies neuronal discontinuity. Put differently, stimulus-induced activity can be regarded a discontinuous neuronal extension of the brain’s resting-state activity.

This suggests that resting-state and stimulus-induced activity differ only in degree, not in principle. I therefore suggest that the distinction between resting-state and stimulus-induced activity is a matter of degree rather than a matter of principle.

Neuroempirical Remark IIB: “Matter of Degree” Versus “Matter of Principle”

Based on the neuronal continuity, one would suggest that the same kind of neuronal effects that can be observed during stimulus-induced activity should also in principle be possible during resting-state activity; this should be the case in those instances when rest–rest interaction exhibits the same degrees—that is, differences in the diverse measures—that are usually rather associated with rest–stimulus interaction. Let me specify this further in the following.

I demonstrated rest–stimulus interaction during visual or auditory perception to go along with strong activity changes in visual or auditory cortex (see Chapter 11). If the resting state itself, that is, rest–rest interaction, now shows equally strong activity changes in, for instance, auditory cortex, one would expect analogous behavioral and phenomenal states to occur. (p.304)

This is indeed supported by empirical evidence. For instance, auditory hallucinations in schizophrenia can be characterized by abnormally strong and large rest–rest interaction in auditory cortex, which is then associated with the same phenomenal-perceptual state, the perception of voices, as the same degree of change induced by an external stimulus, a real voice. (see Chapter 22 for details, as well as Northoff and Qin 2011; and Northoff 2011).

Another instance may be dreams. We perceive an external world in our dreams despite the fact that we sleep. And, as in the awake state, we experience emotions and a sense of self. One may consequently propose that the rest–rest interactions in the dreaming state show as strong degrees as the ones during rest–stimulus interaction in the awake state; the rest–rest interaction, then, no longer functions as mere rest–rest interaction but rather as “rest–as-if stimulus interaction” (see Chapter 25 and especially Chapter 26 for details, as well as Northoff 2011). This reaches deeply into the realm of consciousness and is therefore delegated to Volume II (see Part VII in Volume II).

These and various other examples (see Chapters 25 and 26 in Volume II for more details) support my assumption of resting-state and stimulus-induced activity not being principally different. If they were principally different, neither of the behavioral and phenomenal states associated with stimulus-induced activity could possibly be elicited in the resting state itself; that is, during rest–rest interaction. If so, the neuronal difference between resting-state activity and stimulus-induced activity cannot be a principal difference: it is not a “matter of principle” but rather a “matter of degree.”

Neuroempirical Remark IIC: “Priority of Degree and Difference” Versus “Priority of Origin and Stimulus”

How can we further specify the concept of “matter of degree”? The concept of “matter of degree” describes mere statistical differences, for example, the statistical frequency distribution of neural activity changes across different discrete points in physical time and space. Stronger activity changes are usually associated with rest–stimulus interaction as the interaction between the stimuli’s natural statistics and the resting state’s neuronal statistics, whereas weaker activity changes occur normally during rest–rest interaction and thus within the resting state’s neuronal statistics itself. Accordingly, resting state and stimulus-induced activity are distinguished on the basis of mere statistical differences signifying a “matter of degree” between the resting state’s neuronal statistics and the stimuli’s natural statistics.

Reliance on mere statistical differences implies that the origin of the stimuli, intero- or exteroceptive or neural, is secondary in determining the associated neuronal, behavioral, and phenomenal states. Any stimulus of whatever origin, whether interoceptive, exteroceptive, or neuronal, can in principle induce and elicit the kind of strong neural activity changes that are usually associated with exteroceptive stimuli. This is so because the resulting neural activity is not primarily based on the origin of the stimulus (see Chapter 12 herein and especially Chapter 25 in Volume II for more details on this point).

Instead, neural activity is based on the degree of the statistically based spatial and temporal differences the stimulus introduces (relative) to the brain’s ongoing resting-state activity. If the difference is large, indicating strong rest–stimulus interaction, large neural activity changes will be elicited and accompanied by strong behavioral and phenomenal effects. If, in contrast, the difference is rather small, indicating weak rest–stimulus interaction, the stimulus’ effects will be small, too.

Since neural activity is encoded and determined on the basis of statistically based spatial and temporal differences, any stimulus of whatever origin can in principle elicit any kind of stimulus-induced activity, including its associated behavioral and phenomenal (and psychological and mental) effects.

One may consequently speak of “priority of degree” as a “priority of differences” to characterize the relationship between resting-state and stimulus-induced activity. In contrast to the degree of difference, the origin of the stimulus remains secondary: the origin of the stimulus only matters if it leads to statistically based (p.305) spatial and temporal differences, whereas the origin itself, independent of associated statistical differences, does not matter for determining the degree of neural activity. There is thus no “priority of origin” as a “priority of stimuli” in the brain’s resting-state and stimulus-induced activity. There is “priority of difference and degree” rather than “priority of stimulus and origin” in rest–stimulus interaction.

Neuroempirical Remark IID: Resting-State Activity as Neural Predisposition of Stimulus-Induced Activity

I demonstrated that stimulus-induced activity is dependent upon the resting-state activity, as is well reflected in the assumption of a neuronal continuum. However, the neuronal continuum goes hand in hand with a neuronal discontinuum. This means that the resting-state activity is not sufficient but only necessary for stimulus-induced activity.

The resting-state activity is therefore what I describe as the “neural predisposition” for subsequent stimulus-induced activity. Within the present context, the term “neural predisposition” refers to the necessary but not sufficient neuronal conditions of stimulus-induced activity (see Introduction in Volume II for further discussion of the concept of “neural predisposition, ” as well as Northoff 2013).

As a neural predisposition, the resting-state activity determines the possible and thus available ranges of the degree of subsequent stimulus-induced activity. This was, for instance, indicated in that the resting-state activity sets the ranges for the possible degrees of difference-based coding during subsequent stimulus-induced activity (see Chapter 11). By showing, for instance, strong degrees of functional connectivity in the resting state, the ability of the subsequent stimulus to further increase the degree of functional connectivity is limited, whereas the opposite is the case if the resting state’s functional connectivity is rather weak. The resting state thus predisposes the range of possible options that the subsequent stimulus-induced activity can possibly take (see Fig. A1-1b).

Neurometaphorical Excursion IA: Brain and Supermarket

Let us illustrate the relationship between resting state and stimulus-induced activity by the analogous and metaphorical example of shopping in a supermarket. The supermarket displays various products. What you actually buy depends very much on your own budget and your mood and how that meshes with the products displayed. If you have plenty of money, you may go to the more expensive products. If your mood is gloomy, you may avoid the colorful and shiny products; and so forth. What you actually buy can thus be traced back to what one may want to call “supermarket–customer interaction.” Needless to say, that corresponds very well to what I described as rest–stimulus interaction in the brain.

Now, let us assume that the supermarket happens to be in a neighborhood that recently changed considerably, with many rich people moving in. Naturally, these people look for more high-quality high-priced products. The supermarket’s previous strategy of offering more low-quality and low-priced products may need to change, considering the neighborhood’s influx of rich people. Hence, the supermarket may shift its focus and adapt its products to the new clients by displaying more high-quality and high-priced products. There is thus what can be described as “customer–supermarket interaction.” Needless to say, this corresponds well to stimulus–rest interaction in the case of the brain.

Where, though, does the resting state’s neural predisposition find its analogue in our example of the supermarket? The supermarket is characterized by certain spatial and temporal features; its building is rather small, and everything is extremely tight. These are the constraints within which the shift in focus, from low-quality to high-quality products, can take place. Beyond that, nothing is possible.

This means that, due to the smallness of its boards, shelves, and display tables in the overall extremely tight space, big high-priced products cannot be displayed, meaning that customers interested in these will not find anything in the supermarket. The supermarket’s spatial (p.306) and temporal structure (its shelves, boards, and tables) thus provides the ground upon which the products can be selected and the customers that can possibly be attracted. In other words, the supermarket’s spatial and temporal features provide the predisposition for the range of possible options for subsequent products and customers.

Neurometaphorical Excursion IB: Supermarkets and Consciousness

Needless to say, the supermarket’s spatiotemporal predisposition for certain types of products and customers corresponds to the resting state’s spatiotemporal structure, as it predisposes the resting state to process (weaker or stronger) particular stimuli. As the supermarket’s spatiotemporal structure allows for certain opportunities and prevents others by means of its spatiotemporal features, so does the brain’s resting state and its spatiotemporal structure provide what I described as “spatiotemporal window of opportunity” (see Chapter 11).

The resting state’s “spatiotemporal window of opportunity” can thus be characterized as a neural predisposition; that is, a necessary but not sufficient condition, of possible stimulus-induced activity. In addition to its central importance for subsequent stimulus-induced activity, I propose that the resting state’s “spatiotemporal window of opportunity” also provides the neural predisposition for the behavioral and phenomenal states associated with the stimulus-induced activity.

More specifically, this means that I consider the resting state and its spatiotemporal structure to be a neural predisposition of possible consciousness (NPC) (which as such must be distinguished from what is currently discussed as neural correlates of consciousness [NCC]). This reaches deeply into the realm of consciousness and will therefore be delegated to Volume II.

How does the brain’s predisposition for consciousness relate to our example of the supermarket? This is the point where my analogy finally breaks down, with brain and supermarket parting from each other. In contrast to the brain, supermarkets will never be able to provide a predisposition for consciousness. Accordingly, to put it succinctly, brain is not supermarket and supermarket is not brain. Aren’t we lucky that we are owners of a brain that allows us to create supermarkets (on the basis of our consciousness) rather than being owners of a supermarket that (can only) create(s) brains (without consciousness)?