2. The role of feedback in language learning

Natural language acquisition and learning motivates a psychological meaning for feedback competence, described as the knowledge of the results of any behavior, considered as influencing or modifying further performance (Webster’s Encyclopedic Dictionary of the English Language, 1989). Since knowledge can be understood as information gained via experience or study, the present framework is capable of the psychological reference.

As a pursuit of patterns of behavioral validity, language acquisition is language learning, where knowledge of result may correspond with returning of output, yet human individuality might impel a revision on feedback as return to part own input. Closed-loop competence to become affirmed for all stages of linguistic progression, a term of human feedback strategy shall be proposed.

2.1. Language within a program perspective

Linguistic studies have noted on a tendency to position speech and language within a perspective for human “programming” of behavior, own or by another organism. Wiktor Jassem presents an indication by Paul Postal (1987):

“…a language is an infinite set of sentences which are triplets of phonetic, syntactic, and semantic properties generated by a finite abstract project, or grammar, which consists of sets of partially independent elements called rules and a lexicon or dictionary. Such grammars are represented in human neural systems and provide implicit knowledge of the language they define. A grammar is thus in certain ways analogous to a computer program in that it is a formal system partially determining the behaviour of a physical system (…)”

Capacity by language to “program” behavior would belong with a different scope of inquiry, also as declinable. Feedback to remain the focus of the present work, two prominent approaches to language development can be quoted.

The nativist party posit that humans are born with DNA-encoded language universals. Nevertheless, Noam Chomsky has stated for his language acquisition device to depend on experience as releaser, as well as on exposure for perfection. The nativist tenets point to insufficiency of environmental variables and the naturalness of the postulated generative universals, discerned for unrelated languages (Akmajian et al., 1985).

Cognitive linguists insist on mental capabilities to precede language refinement (ibidem). Intellectual progress would be contingent on brain maturity as well as personal intellection, the cognitivist sense for intellective advancement to attend to capability to extract from experience and build heuristics. The approach places language learning in context with problem solving.

Nativism or cognitivism, it is reasoned that both need feedback phenomena. The present argument selectively does adopt the neoconnectionist approach and memory network models. Attempts to declare a brain singular site for human memory have been failures without exception. Bruce Derwing’s guideline (in Jassem, 1987) that “no other, special mechanisms or secret abilities are necessary for learning language than learning anything else”, is the principium of the inquiry.

2.2. The closed-loop process of neural network forming

Biological maturation alone is unable to explain linguistic progress, though specificity increase does enhance neural integrity. Learners may never attain linguistic finesse without favorable exposure to embrace standard spoken and written resources as in libraries and media. Two major approaches to human learning, behaviorism and connectionism, differ fundamentally.

Ivan Pavlov postulated that learning was substantiated by brain reflexes. He ascribed reflex forming to temporal co-occurrence of events. Pavlov’s approach, often misjudged for an associationist venture (Szewczuk, 1984), was opposed by American connectionists, who insisted on personal recognition of result or response as a competence of intrinsic affective quality. The laws of learning by Edward Lee Thorndike were those of exercise, readiness, and effect.

Neural schemata are basic biological forms for natural language (Puppel, 1992), and the laws can be compared with the feedback model for relevant pattern forming. Respectively, exercise would increase pathway use; effect affirming or denying on validity, neural readiness would encourage rehearsal of connectivities to have produced agreeable consequents. Literature in psychology proceeds with augmented attention to neural “labels” for affect, owing to the role in path priority (Goleman, 1997).

Clark Leonard Hull (1940, in Szewczuk, 1984) defined need as a negative balance between the individual and his or her setting. A behaviorist signaling value would target the individual, either to accommodate the inner equilibrium, or to threaten it. Induced co-occurrence of events and forced response would be allowed in the context as well. The role of concomitant conditions came to be stressed also by Edwin Ray Guthrie (1935, in Szewczuk, 1984). Thorndike questioned the worth of simultaneity, insisting the affective component was inherent to all learning.

For standard as well as school contexts, recognition between an impulse or mere presence is by and large individual. In neural path forming, signal objective incidence is not likely to prevail over the favored signaling quality, especially for educated and fluent language styles, where personal congruity is the predominant motive.

Therefore, a closed-loop schema may represent neural patterning for language, human inner actuation always to refer for affective congruity: attainment at undesirable objectives or failure at acceptable skill, both entwine with affective disadvantage. Affect is not understood for emotion alone, since it does not have existence without consciousness and perception.

Figure 2. A general, closed-loop schema for neural patterning. S — Signal; A — Inner actuation state to embrace affective congruity; S1 — Memory reflection on signal; S2 — Processed signal inner representation; R1 — Response; R2 — Result.

Figure 2 is to connote actual neural connectivity, as concepts on purely “functional” or “mathematical” significs are of no meaning to biological systems (Szewczuk, 1984). The generalization is not to commend an oversimplified view of human neural links. Patterns and schemata work together, the function in network intermediate layers to escape monitoring in continual, compound, and intricate dynamics. Figure 2 constituent nodes are to symbolize multi-layered structures able to act as in Figure 3 (in Puppel, 1992).

The schema of Figure 2 involves memory reflection on signal, whereas there are no memory precedents for novel phenomena. Standard language competence usually yet affords verbally to define new percepts, allowing new perspectives for familiar phenomena as well.

Affinity and chemical specificity may account for encoding processes only in a proportion. It would be hypertonic backlash (Szewczuk, 1984) or neuron differential tonicity, to mediate path validation. As predicted in Thorndike’s law of readiness, the tonicity would ensue from synergisms among existent schemata, the process to require intrinsic feedback. Figure 4 presents a generalization for hypertonic backlash as memory pattern reinforcer.

Figure 4. Memory natural backlash reinforcement. S — Signal, R — Response within the neural network, B — Backlash reinforcement (compare Szewczuk, 1984).

Natural neural chemistry and voltage operate on minim quantities, which denies validity to behaviorist approaches. Individuality effects within standard neural function have become of neoconnectionist focus.

2.3. Network feedback-mediated functioning

Networks are biologically standard structures, humans to be potentially unique in pattern integration also for behaviors everyday or formal. Personally efficient convergence has been proposed for the primary formative in encoding, inactive information to be subject to forgetting (Szewczuk, 1984). Network layering has been classed in two types, the symbolic and the sub-symbolic connectivity. The sub-symbolic layers would merge for praxic and motor adeptness, the symbolic networks to build declarative scopes for intellectual routines and creativity (Puppel, 1992).

Network dynamics is mostly stochastic or deterministic in manner. The random mode may be observed intrapersonally, with volitional fixation of gaze and motor as well as cognitive inactivity. Brain natural exercise on existent linkage, the mode may add to network readiness. A highly deterministic network behavior can be recognized about skilled and proficient language uses, taken the cerebral specialization in human thought, memories, imagery, and linguistic performance.

Reservations on response as different from signal may arise, whereas “returning of signal” or “signal reiteration” have been in use to define feedback as a process. The cerebellum is known to connect with the neocortex. Motor command change or reset requires signal variance, yet the connection is of mutual effectiveness within ongoing motor sequences. By the criterion, inner feedback would be a closed-loop process between neural structures active in the same real-time sequence.

Cerebellar communication is outside interoception. Feedback reliance in instinctive human behavior deserves inquiry for human language acquisition and learning.

2.4. Circular reactions in child development

In psychology and physical education, three natural continua for human ontogenetic progression have been the most acknowledged, the motor, the affective, and the cognitive activity. The three types of human pursuit have been found in interaction and rapport as to allow the term “psychomotor”, with regard to human locomotion and language behavior (Payne and Isaacs, 1987).

The term is not to conclude on the influence as top-down strictly. Motor exercise has been proved an asset to cognitive ableness, the partial intellectual decline theory to have named motor regression for a factor in advanced adulthood cognitive slowdown (ibidem). For speech and language, the term “psychomotor” is to emphasize the necessary priority of the human person, in speaking and writing.

The notions of language development rate and order also invite comment, as in Crystal, Fletcher, and Garman (1976). Specifically, language acquisition order would show statistic regularity, humans potentially to vary far more in language maturation pace, the order and rate thus to make distinct variables.

The present projet for language refinement builds on the notion of circular reactions as noted by Jean Piaget (1992) about infant instinctive activity. The child would introduce changes into a setting, watch the course of alterations effected, and internalize cognitively pertinent variables. Piaget described a sensorimotor infant as “thinking by bodily movement”. The child would be absorbed with practice of own kinesthetic range up to one, or one and a half months of age.

Since this is the human individual to make as well as cerebrate on own variables, circular reactions are closed-loop behaviors, the human individual to be the actual, own focus. The natural onset to belong with the sensorimotor stage of human life, the closed-loop behaviors must be biologically motivated. Until the fourth month, the child would repeat own earliest movements, watch and internalize their course, which has been classed as primary circular reactions.

Secondary circular reactions are primary patterns modified for new circumstances. The child learns to combine own initial behavior, prolonged workout to integrate sensory awareness and motor command. The secondary schemata help improve movement coordination.

Around twelve months of age, the interplay of sensory, motor, and cognitive powers becomes tertiary reactions. Trial and error to have encouraged prediction beyond immediate manipulation of objects, the child begins to show interest in the environment. In object handling, attempts to act through others inspire socializing; objective localization gradually to replace egocentric spatial awareness, coincidence-anticipation timing grows into directional cognizance. The skill later acts in such academic tasks as reading (Payne and Isaacs, 1987).

The human individual to make variables within own neural structures, the circular rehearsal would require some return to own working input. A view to inner executive controls can help discern on feedback as potentially elaboration on inputs.

2.5. The executive controls theory by Robbie Case

Robbie Case, a psychologist, attempted to merge Jean Piaget’s approach with own information processing model. He posited that new executive controls made for increased intellectual means in cognitively refining children. The concept of new controls was to denote “the current ways of representing the world coupled with the child’s current strategies with specific kinds of problems” (Wortman et al., 1988).

The researcher evaluated children’s intellection and short-term memory by two tasks he assumed of the same logical insight, quantity evaluation by number or volume. He found competence in estimating number and deficit in cognizance on volume a regular pattern among six-year-olds. He premised an executive controls discrepancy he named a horizontal décalage (ibidem).

Human short-term memory is speculated not to change past maturation (Wortman et al., 1988); the name “working memory” has been in use for either the short-term capability, or a hypothetical inventory to mediate between the short and long term. The working performance has been exposited as potentially labile (Kurcz, 1992). The matter of executive controls might be that of neural network efficiency.

All languages discern between number and mass, countability to precede the grammatical article, in acquisition of French as well as English. Grammatical gender is another connotation, for French and other languages that assign gender. Therefore, the “horizontal décalage” might suggest a lag where there is not delay: mastery of grammatical determinateness is not expected of the six-year-old.

Mass or volume relate to determinateness and require neural linkage broader than that for telling quantity alone. The “décalage” would narrate on a standard developmental stage where neural coherence is only yet not sufficient. Inner integration to require own inner inputs, a feedback strategy rather than new controls might be recognized, about human language acquisition and learning.

2.6. Feedback exercise in child language learning

Most children show extensive aptness in speech sound imitation, linguistic analogy, and inductive reasoning. Language development begins with instinctive vocalizing or cooing, the activity to correspond in consequence with infant sensorimotor coaptation. Children usually have response from other humans, the earliest “proto-conversations” to help perceive phonation as feedback interaction, already in the first year of life.

The one-word stage would compare with primary circular reactions: the child repeats single words or sound clusters, the worth of response to be not only interaction or need fulfillment, but also vocal pattern validity. Utterance rich interpretation and paraphrase may embolden progress into the “gimme” or “allgone” units of meaning as by Crystal (1976). The verbalizations merge one-word patterns for language behavior, and a parallel between them and secondary circular reactions can be posited.

A strategy may emerge between ages two or three. Asking Why-questions, when the answer is known, only the context in which to ask not fully mastered, may exemplify a developing strategy. The language standard always to remain beyond arbitrary manipulation, the repetitiveness may reflect on success with language tertiary reactions, the often “telegraphic” styles to have gained grammar via recurrent restructuring. Humans may vary remarkably in strategies for intonation, syntax, vocabulary, as well as interlocutor choices (ibidem).

Environmental closed-loop linguistic activity as in Figure 1 would begin to take shape already with language secondary circular reactions. Dependence on environment would later diminish, the closed-loop manner of egocentric speech usually to assist progress towards linguistic and intellectual sovereignty.

2.7. The closed-loop behavior of egocentric language

Egocentric speech occurs spontaneously, in children two to seven years old. Children rehearse locution mostly with regard to own behavior, being themselves the intended perceivers of the spoken exercise. Egocentric talk strengthens procedural thinking. Playing or resolving on a task, the child may comment on concurrent as well as anticipated events, verbalization to expedite activity planning and modeling (Jurkowski, 1986).

In spoken form, self-oriented practice helps shape phono-articulatory patterns. Written egocentric exercise would follow, to facilitate the neuromotor consecution for handwriting. However, language sequencing never is the ultimate goal of maturational activity: spoken or written, language practice encourages integration of memory declarative and procedural paths (ibidem).

Intrapersonal communication may continue throughout lifespan. In note making, the author of the written message is the intended reader, specific or cursory character to make the message potentially unintelligible to other people (Pazukhin, 1996/97). The lifespan prospect allows to posit a generally feedback pattern in human learning and skill.

2.8. The generally feedback pattern in human learning and skill

Neural schemata may recede without systemic actuation, as in language users to lose fluency in desuetude. Style modeling to remain possible past maturation, local as well as central feedback would always be part in human neural structuring and upkeep. Everyday living necessitates ability to manage with novel circumstances and information: growth, exception, and change of synaptic significance are part every human life (Vander et al., 1985).

Learning involves feedback-mediated neural link expansion, as well as path delimitation. Neural connectivity extends for initial stages of a cognitive process, further phases to bring connectedness reducing within personal neural economy (Kurcz, 1992). Biological concision allows considerable preservation of knowledge without much spatial requirement on the brain.

The time envelope to be very strict for single neurons, experiments with conditioned learning have showed a major tolerance for reinforcement delay, in complex cognitive processes (Kurcz, 1992). Conditioning to postulate program perspectives for intellection, the relaxed time frame might have come with ascent in human inner hierarchy to involve enhanced feedback capabilities.