The article introduces a new theory of subjective probability called Support Theory, which posits that different descriptions of the same event can lead to different judgments. The theory is based on the idea that probability judgments are influenced by the explicitness of the descriptions rather than the events themselves. Key features of Support Theory include: 1. **Nonextensionality**: The theory relaxes the extensionality principle, which states that events with the same extension should have the same probability. Instead, it allows for different descriptions of the same event to produce systematically different judgments. 2. **Unpacking Principle**: Judged probability increases when the focal hypothesis is unpacked and decreases when the alternative hypothesis is unpacked. This principle is consistent with experimental evidence. 3. **Subadditivity**: Judged probabilities are subadditive for implicit disjunctions, meaning that the probability of an implicit hypothesis is less than the sum of its components. This is in contrast to classical and revisionist models of belief. 4. **Enhancement**: The theory predicts that evidence that is more compatible with the hypotheses under study will induce greater subadditivity. 5. **Ordinal Generalization**: The theory can be extended to ordinal judgments, where probability judgments are treated as ordinal rather than cardinal. 6. **Conditional Probability**: The theory provides a representation of conditional probability, showing how new evidence can revise the support function. The article reviews experimental evidence supporting these predictions, including studies on the unpacking effect, binary complementarity, and subadditivity in both laypeople and experts. The findings suggest that Support Theory offers a unified account of various empirical phenomena and provides new insights into the nature of subjective probability.The article introduces a new theory of subjective probability called Support Theory, which posits that different descriptions of the same event can lead to different judgments. The theory is based on the idea that probability judgments are influenced by the explicitness of the descriptions rather than the events themselves. Key features of Support Theory include: 1. **Nonextensionality**: The theory relaxes the extensionality principle, which states that events with the same extension should have the same probability. Instead, it allows for different descriptions of the same event to produce systematically different judgments. 2. **Unpacking Principle**: Judged probability increases when the focal hypothesis is unpacked and decreases when the alternative hypothesis is unpacked. This principle is consistent with experimental evidence. 3. **Subadditivity**: Judged probabilities are subadditive for implicit disjunctions, meaning that the probability of an implicit hypothesis is less than the sum of its components. This is in contrast to classical and revisionist models of belief. 4. **Enhancement**: The theory predicts that evidence that is more compatible with the hypotheses under study will induce greater subadditivity. 5. **Ordinal Generalization**: The theory can be extended to ordinal judgments, where probability judgments are treated as ordinal rather than cardinal. 6. **Conditional Probability**: The theory provides a representation of conditional probability, showing how new evidence can revise the support function. The article reviews experimental evidence supporting these predictions, including studies on the unpacking effect, binary complementarity, and subadditivity in both laypeople and experts. The findings suggest that Support Theory offers a unified account of various empirical phenomena and provides new insights into the nature of subjective probability.