Adaptability and the integration of computer-based information processing into the dynamics of organizations

The structure of a system influences its adaptability. An important result of adaptability theory is that subsystem independence increases adaptability [Conrad, M., 1983. Adaptability. Plenum Press, New York]. Adaptability is essential in systems that face an uncertain environment such as biological systems and organizations. Modern organizations are the product of human design. And so it is their structure and the effect that it has on their adaptability. In this paper we explore the potential effects of computer-based information processing on the adaptability of organizations. The integration of computer-based processes into the dynamics of the functions they support and the effect it has on subsystem independence are especially relevant to our analysis.     

Biological information processing: the use of information for the support of function

In biological systems, the processing and use of information has evolved out of the need for survival in the face of an uncertain environment. As a consequence, the information-function relationship in these systems is shaped by their adaptability characteristics. In contrast, the information-function relationship in man-designed, goal-oriented organizational systems depends on the ability of the information processing system to support the achievement of the organization's goals. In this paper we use results from adaptability theory in the analysis of control-related aspects of the information-function relationship in man-designed organizational systems. In particular, we use a conceptual model of organizational control to characterize features of functional and control structures and their effect on the adaptability of these systems. The concept of implicit control and a design principle for adaptability-enhancing information systems are derived for this analysis.     

Digital and biological computing in organizations.

Michael Conrad unveiled many of the fundamental characteristics of biological computing. Underlying the behavioral variability and the adaptability of biological systems are these characteristics, including the ability of biological information processing to exploit quantum features at the atomic level, the powerful 3-D pattern recognition capabilities of macromolecules, the computational efficiency, and the ability to support biological function. Among many other things, Conrad formalized and explicated the underlying principles of biological adaptability, characterized the differences between biological and digital computing in terms of a fundamental tradeoff between adaptability and programmability of information processing, and discussed the challenges of interfacing digital computers and human society. This paper is about the encounter of biological and digital computing. The focus is on the nature of the biological information processing infrastructure of organizations and how it can be extended effectively with digital computing. In order to achieve this goal effectively, however, we need to embed properly digital computing into the information processing aspects of human and social behavior and intelligence, which are fundamentally biological. Conrad's legacy provides a firm, strong, and inspiring foundation for this endeavor.     

The mutation buffering concept of biomolecular structure

Redundant elements in proteins and nucleic acids serve to buffer the effect of point mutations on features of conformation critical for function. Mutation buffering associated with mechanistically redundant amino acids facilitates the evolution of proteins. Such redundant amino acids accumulate by hitch-hiking along with the evolutionary advances which they facilitate. Redundancies in DNA (such as introns and repetitive DNA) prevent extraneous sequence dependent conformational effects from interfering with readout. They also facilitate regulatory evolution. According to the mutation buffering concept biological organizations are selected to facilitate evolution. As a consequence biological information processing is very different from information processing in man-made computers. The link between molecular conformation, evolutionary processes, and information processing is formulated in terms of a tradeoff principle. By utilizing mutation buffering biological systems sacrifice programmability; by achieving programmability digital computers make mutation buffering computationally expensive and hence sacrifice evolutionary adaptability.     

Bootstrapping on the adaptive landscape.

Different versions of a gene or of a multigenic system may be essentially equivalent so far as the specific function of the structures which they code for or control is concerned, but very different with respect to their amenability to evolution. The structural features which increase evolutionary amenability are a disadvantage to the organism in terms of energy. Nevertheless, they accumulate in the course of evolution as a consequence of hitchhiking along with the desirable traits whose evolution they make possible. This is the bootstrap principle of evolutionary adaptability. In terms of the adaptive landscape bootstrapping corresponds to populations evolving in such a way that they occupy regions of the landscape which are more amenable to evolutionary hill climbing. The bootstrapping idea has implications for structure-function relations in a number of complex biological information processing systems, including biochemical systems, the immune system, and the brain. Bootstrapping is also discussed in connection with the origin of information processing (the origin of life) and in connection with possible designs for macromolecular computing systems.     

Complex adaptation and system structure

The structural organization of biological systems is one of nature’s most fascinating aspects, but its origin and functional role is not yet fully understood. For instance, basic adaptational mechanisms like genetic mutation and Hebbian adaptation seem to be generic and invariant across many species and are, on their own, fairly well investigated and understood. However, it is the organism’s structure – the representations these mechanisms act upon – that bears the complex functional effects of these mechanisms. While typical technical approaches to system design require detailed problem models and suffer from the need to explicitly take care of all possible cases, the organization of biological systems seems to induce inherent adaptability, flexibility and robustness. In this discussion paper we address the concept of structured variability, particularly the role of system structure as implementing a certain representation on which basic variational mechanisms act on. The functional adaptability (or search distribution) depends crucially on this representation.     

Visually guided gait modifications for stepping over an obstacle: a bio-inspired approach

There is an increasing interest in conceiving robotic systems that are able to move and act in an unstructured and not predefined environment, for which autonomy and adaptability are crucial features. In nature, animals are autonomous biological systems, which often serve as bio-inspiration models, not only for their physical and mechanical properties, but also their control structures that enable adaptability and autonomy—for which learning is (at least) partially responsible. This work proposes a system which seeks to enable a quadruped robot to online learn to detect and to avoid stumbling on an obstacle in its path. The detection relies in a forward internal model that estimates the robot’s perceptive information by exploring the locomotion repetitive nature. The system adapts the locomotion in order to place the robot optimally before attempting to step over the obstacle, avoiding any stumbling. Locomotion adaptation is achieved by changing control parameters of a central pattern generator (CPG)-based locomotion controller. The mechanism learns the necessary alterations to the stride length in order to adapt the locomotion by changing the required CPG parameter. Both learning tasks occur online and together define a sensorimotor map, which enables the robot to learn to step over the obstacle in its path. Simulation results show the feasibility of the proposed approach.     

The seed germination model of enzyme catalysis.

The activity of enzymes and other biological macromolecules is often sensitively dependent on physiochemical context. Seed germination provides an analogy that helps to elicit the control and information processing capabilities of enzymatic networks. Like a seed, the enzyme takes a particular action (complexes with a specific substrate and catalyzes a specific reaction) when a specific set of milieu influences is satisfied. The context sensitivity, specificity and speed are enormously enhanced by the parallelism inherent in the electronic wave function (i.e. by the superposition of electronic states). This parallelism is converted to speedup through electronic-conformational interactions. The quantum speedup effect allows biological 'switches' to have qualitatively greater pattern recognition capabilities than electronic switches. Consequently the information processing and control capabilities of biomolecular systems exceed the capabilities obtainable from classical models and exceed the intuitive expectations that have developed through the study of such models.     

Biologically Inspired Information Processing and Synchronization in Ensembles of Non-Identical Threshold-Potential Nanostructures

Nanotechnology produces basic structures that show a significant variability in their individual physical properties. This experimental fact may constitute a serious limitation for most applications requiring nominally identical building blocks. On the other hand, biological diversity is found in most natural systems. We show that reliable information processing can be achieved with heterogeneous groups of non-identical nanostructures by using some conceptual schemes characteristic of biological networks (diversity, frequency-based signal processing, rate and rank order coding, and synchronization). To this end, we simulate the integrated response of an ensemble of single-electron transistors (SET) whose individual threshold potentials show a high variability. A particular experimental realization of a SET is a metal nanoparticle-based transistor that mimics biological spiking synapses and can be modeled as an integrate-and-fire oscillator. The different shape and size distributions of nanoparticles inherent to the nanoscale fabrication procedures result in a significant variability in the threshold potentials of the SET. The statistical distributions of the nanoparticle physical parameters are characterized by experimental average and distribution width values. We consider simple but general information processing schemes to draw conclusions that should be of relevance for other threshold-based nanostructures. Monte Carlo simulations show that ensembles of non-identical SET may show some advantages over ensembles of identical nanostructures concerning the processing of weak signals. The results obtained are also relevant for understanding the role of diversity in biophysical networks.     

Limits on the computing power of biological systems

The theory of computational complexity and certain explicitly-stated hypotheses imply limitations on the information processing power of biological systems. Parallelism, special purpose organization, and analog mechanisms may provide speedup critical for life processes, but have little power in the face of exponential growth. We show that “polynomially simulatable” biological systems cannot exhibit dynamic behavior which produces the solution of an intractable problem. The argument implies that parallelism does not allow biological systems to defeat the exponential explosion, but rather is important because it allows polynomial time algorithms to be used more efficiently.     

Self-control: Refinement of a construct

Previous efforts, on a theoretical/model building level, refined the construct “self-control” into four quadrants: (1) positive assertive (active control), (2) positive yielding (letting-go control), (3) negative assertive (over-control), and (4) negative yielding (too little control). To test the discreteness of the four quadrants, 706 individuals, the majority in health and healing professions, from nine cities across the United States, responded to prompt words designed to assess each quadrant. A factor analysis provided partial concurrent validation, and the results revealed information about the semantic structure of self-control, as well as personal characteristics associated with self-control. Further, mean tabulations showed not only cultural bias (i.e., high self-control was associated almost exclusively with Quadrant 1, but also sex role bias (i.e., low self-control for a man was most often associated with Quadrant 3, negative assertiveness, and for a woman was most often associated with Quadrant 4 negative yielding). Clinical implications of these findings in terms of developing a self-control assessment inventory for matching self-control strategy to an individual with a particular clinical problem are discussed, and guidelines and suggestions for further research are offered.     

Feedback regulation of EGFR signalling: decision making by early and delayed loops

Human-made information relay systems invariably incorporate central regulatory components, which are mirrored in biological systems by dense feedback and feedforward loops. This type of system control is exemplified by positive and negative feedback loops (for example, receptor endocytosis and dephosphorylation) that enable growth factors and receptor Tyr kinases of the epidermal growth factor receptor (EGFR)/ERBB family to regulate cellular function. Recent studies show that the collection of feedback regulatory loops can perform computational tasks - such as decoding ligand specificity, transforming graded input signals into a digital output and regulating response kinetics. Aberrant signal processing and feedback regulation can lead to defects associated with pathologies such as cancer.     

The expansion of information in ecological systems: Emergence as a quantifiable state

Although the term ‘emergence’ has received wide attention in the literature, most of this attention has been focused on epistemological discussions about the nature of what might be considered emergent behavior in self-organizing systems. For the concept of emergence to have any great utility for biologists, it must (1) be perceptible as a physical, quantitative property rather than just a philosophical one; (2) have a quantitative definition applicable to all levels of biological organization; and (3) be an essential component of biological system performance or evolution. Using an independent, cellular population model (running in the StarLogo system), we have developed a mutual information calculation to measure the information expansion when considering the interactions between a population of herbivores and an environment in comparison to the interactions between the individual herbivores and that environment. In self-organizing biological systems, the collective action of massively parallel units generates a greater potential complexity in the information processing capacity of the ‘whole’ system relative to the ‘individual’ parts, and as such, there is a demonstrable increase in mutual information content. From this perspective, we consider emergence to exist as a simple information expansion that is a default behavior of any system with multiple, component parts governed by a simple, probabilistic rule set. It is not a first principle of self-organizing biological systems, but rather a collective behavior that can be quantitatively described in practical terms for experimental biologists. With a quantitative formulation, the concept of emergence may become a useful information statistic in assessing the structure of biological systems.     

Information processing in molecular systems

Information processing, or selective dissipation, is mediated by switching elements in classical systems and by enzyme catalysis in biochemical systems. There are important differences in the character of this dissipation (from the standpoint of energy and control) in self-reproducing systems based on molecular interactions and those based on conventional computers. Conventional computers process information in a single level mode, i.e., the state of each unit of the system is accessed independently. This includes the manipulable memory units which store the computer program. In contrast, molecular self-reproducing systems process information in a hierarchical mode, based on the fact of hierarchy in molecular structure. In particular, the enzyme is described genetically at the primary level of structure (amino acid sequence) but functions at the higher, tertiary level on the basis of the interactions of many manipulable units. As a consequence it is not possible to program a biochemical system in any conventional sense. However, this is compensated by an increased capacity for accumulating appropriate information through evolution by variation and natural selection. This is possible because systems operating in the hierarchical mode are amenable to gradual modification of function. The degree of gradualness is itself an evolved property of biological molecules.     

Drug addiction: the neurobiology of disrupted self-control

The nature of addiction is often debated along moral versus biological lines. However, recent advances in neuroscience offer insights that might help bridge the gap between these opposing views. Current evidence shows that most drugs of abuse exert their initial reinforcing effects by inducing dopamine surges in limbic regions, affecting other neurotransmitter systems and leading to characteristic plastic adaptations. Importantly, there seem to be intimate relationships between the circuits disrupted by abused drugs and those that underlie self-control. Significant changes can be detected in circuits implicated in reward, motivation and/or drive, salience attribution, inhibitory control and memory consolidation. Therefore, addiction treatments should attempt to reduce the rewarding properties of drugs while enhancing those of alternative reinforcers, inhibit conditioned memories and strengthen cognitive control. We posit that the time has come to recognize that the process of addiction erodes the same neural scaffolds that enable self-control and appropriate decision making.     

Implicit Associations Have a Circadian Rhythm

The current study shows that people''s ability to inhibit implicit associations that run counter to their explicit views varies in a circadian pattern. The presence of this rhythmic variation suggests the involvement of a biological process in regulating automatic associations—specifically, with the current data, associations that form undesirable social biases. In 1998, Greenwald, McGhee, and Schwartz introduced the Implicit Association Test as a means of measuring individual differences in implicit cognition. The IAT is a powerful tool that has become widely used. Perhaps most visibly, studies employing the IAT demonstrate that people generally hold implicit biases against social groups, which often conflict with their explicitly held views. The IAT engages inhibitory processes similar to those inherent in self-control tasks. Because the latter processes are known to be resource-limited, we considered whether IAT scores might likewise be resource dependent. Analyzing IAT performance from over a million participants across all times of day, we found a clear circadian pattern in scores. This finding suggests that the IAT measures not only the strength of implicit associations, but also the effect of variations in the physiological resources available to inhibit their undesirable influences on explicit behavior.