What is the term for inference

Inferences - understanding texts

Table of Contents

1. Introduction

2. Inference and Coherence
2.1. Bridging inference
2.2. Elaborative inference

3. Discussion


1. Introduction

Understanding of texts (“comprehension”) is a widely researched sub-area of ​​psycholinguistics, which, however, has only received attention for a few decades. When reading a text, for example, the syntax structure must be analyzed and the semantic and morphological content of the utterance decoded. This is followed by inference formations, which put the words together sensibly and make them understandable for us on sentence and text level.

First, it should be explained briefly what is meant by the term “text” in the following.

text(Latin textum: tissue, text) denotes a delimited, coherent utterance in written language, in the broader sense also the non-written, but writable linguistic information.

(de.wikipedia.org/wiki/Text, last change on July 4th, 2009)

This is a written or oral statement that should contain at least two sentences. Texts can of course also take on a much larger scope. Paragraphs such as “the reader” are meant as examples and also relate to the possible listener.

In order to be able to understand utterances, world knowledge is required, which can differ from person to person. The cultural background, level of education, age and personal interests of a person have a very large influence on knowledge of the world. A child in our culture knows about the existence of Santa Claus and an African tribe pays homage to his deity. This knowledge is also called prior knowledge, as it is brought in from the outset in a certain situation and thus mostly simplifies the understanding of a text, but it can also make it more complicated.

It should also be noted that in the literature different terms for individual phenomena can be found, but these have been noted. In the absence of German equivalents, the English terms were retained to be on the safe side.

2. Inference and Coherence

Inferences can be defined in many ways, since the term is not limited to language and represents a basic cognitive principle (Rickheit 2002, p. 72)[1]. There are different formulation approaches, which, however, always deal with the formation of new propositions by linking world knowledge and the newly processed linguistic information from a text in order to create a text unit (Fix / Poethe 2003, p. 214; Ferstl 2006, p iii). “The basis of every inference is existing knowledge” (Rickheit 2002, p. 17), the prior knowledge. The connection of individual sentences leads to conclusions, which often serve the local coherence, i.e. the context of meaning in a specific text (Ferstl 2006, p. 4). Here, the accurate delimitation of the definition and the use of the term inference is qualitatively not entirely clear, as e.g. irony or metaphors go beyond simple conclusions, but are not always clearly involved or excluded from their considerations by researchers (Ferstl 2006, p. 5). In addition, gestures and facial expressions can lead to inferences in a conversation, for example certain hand movements such as rubbing the thumb on the index finger, which can have different meanings on the subject of money, such as “I want to pay!”, “He has a lot Money ”or“ You still owe me money ”- depending on the context.

Another point of discussion deals with the question of the extent to which inferences are formed automatically (Ferstl 2006, p. 22). This question is still to be investigated.[2]

The two terms coherence and inference are inextricably linked and coherence is sometimes referred to as a special type of inference (Traxler 2006, p. 776). Harvey, on the other hand, categorizes this phenomenon as logical inference (Harley 2008, p. 369).

The formation of inference is facilitated by cohesion markings such as pronouns, conjunctions and repetitions (Ferstl, p. 7), since references to previous text units are made automatically. This structure, in which new information meets known information in a sentence, is described by the term theme-rhema. Cohesion markings can, however, make it more difficult to identify breaks in coherence (ibid.).


[1] From a neuroanatomical point of view, it has been proven that the inference formation process takes place to a large extent in the right hemisphere (Ferstl, p. 24; Traxler, p. 777). For example, if there is information in the text that contradicts one's own world knowledge, so-called global inconsistencies, the right temporal lobe is activated (Ferstl, p. Iv).

[2] see chapter 3

End of the reading sample from 10 pages