> For the complete documentation index, see [llms.txt](https://cbontology.gitbook.io/cognitive-bias/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://cbontology.gitbook.io/cognitive-bias/ontologies/general.md).

# General

Considering the project from a broader view, which involves creating an ontology describing the behavior of all cognitive biases outlined in the wheel, we have chosen to develop a generic section of the ontology. This section is designed to gather all the structures of single bias ontologies by formally representing the structure of the wheel shown in[ Overview](/cognitive-bias/readme.md) .

<figure><img src="https://upload.wikimedia.org/wikipedia/commons/6/65/Cognitive_bias_codex_en.svg" alt=""><figcaption></figcaption></figure>

As shown in the picture above, the biases can be grouped around two level of concepts: a broader one in which we have the four concepts that we can see in the conrner of the picture, and a narrower layer where the biases are grouped around more specific concepts. Therefore, we have developed a model to represent this layering of concepts that can be expressed through the following diagram:

<figure><img src="/files/0Lglnpqd1ORHZC4gdXR0" alt=""><figcaption></figcaption></figure>

This model conceptualizes both levels as concepts, using the class Concept of the ODP [Classification](http://ontologydesignpatterns.org/wiki/Submissions:Classification); additionaly, the broader level is defined as Collection of concepts using the ODP [Collection](http://ontologydesignpatterns.org/wiki/Submissions:Collection). Furthermore, we have created a property that links each bias to both the levels, through the property isCausedBy. Also the class BiasedAgent, that represents the agent who is affected by a specific bias, is connected to the class CognitiveBias through the property isAffectedBy; the class NonBiasedAgent, on the other hand, represents other individuals who interact with the BiasedAgent but are not affected by the bias itself. Finally, we have modeled each bias as a class (and not as an individual of the class cognitive bias) since each bias can be described through [sub-categories](https://en.wikipedia.org/wiki/Confirmation_bias#Types).

Link to the OWL file of the General ontology:

{% embed url="<https://raw.githubusercontent.com/giorgiacrosilla/biasontology/main/owl/general.owl>" %}

Final viasualization of all our biases together:

{% embed url="<https://service.tib.eu/webvowl/#opts=doc=0;#iri=https://raw.githubusercontent.com/giorgiacrosilla/biasontology/main/owl/bisso.owl>" %}
