Let us be philosophers! : Computerized support for ethical decision making

Abstract

This thesis presents a computerized tool for ethical decision making. For someone who is unfamiliar with the psychological theory that the tool is based on, it will perhaps first appear as a pointless piece of software. It does not give any guidance to what an ethically correct decision is, it does not suggest relevant ethical principles or guidelines and it does not even make reference to known cases of good moral conduct. In fact, it does not make any moral claims at all. The only two things that the tool does are that it stimulates reflective, analytical and holistic reasoning and blocks automatic, biased and constrained impulses. This approach is chosen to improve the decision maker's ability to consider the relevant circumstances in a situation. By focusing on relevant interests of stakeholders, the scope of consideration in a moral situation can be expanded and the impact of decisions can be evaluated with respect to these. To justify this non-normative approach, the functionality of normative ethics is analyzed. The conclusion stresses the importance of self-conscious deliberation. Further arguments for advocating a systematic, holistic and self-critical handling of moral problems are collected from both philosophy and psychology. The structure and functionality of the tool is founded in psychological theory and especially the problem of cognitive biases in moral decision making is addressed. The tool has been evaluated in two studies, which both indicate that it actually delivers what it was designed to do. Statistically significant results show that the tool helped users to expand the scope of consideration in a moral problem situation compared to using an equivalent paper-and-pen-based method.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 9 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.