從個體臉孔上除了可以得知容貌特徵（性別、年齡等）外，我們還可以經由臉孔推論出他人的社會身分或團體。本論文以「從外在容貌特徵推論他人政黨派別」為主題，以台灣政治環境為背景，探討年齡變項是否會影響政黨身分歸類作業以及特質評分作業的表現。
本論文包含了三個實驗：實驗一為照片知覺年齡評定作業，主要探討國民黨以及民進黨候選人刺激照片的知覺年齡間是否有差異。除此之外，我們也嘗試降低這樣的差異，作為實驗二的刺激照片。實驗二為政黨臉孔照片歸類作業，採用調整過的兩黨候選人照片，進一步以年齡區間分成四組後，探討參與者歸類作業的表現是否會受到影響。實驗三為特質評分作業，主要以六個目標特質，探討國民黨以及民進黨是否有足以將兩黨區分出的特質，以及年輕及中成年參與者是否會在特質評分作業的表現上有所不同。結果發現，就算降低兩黨候選人間照片知覺年齡的差異，參與者在歸類作業中，仍然具有顯著高於機率的表現。此外，隨著照片知覺年齡上升，國民黨候選人正確率也跟著上升。另外，我們也發現省籍可以做為區分兩黨候選人的臉孔特質，但年長參與者還可以用臉孔成熟度進行區分。綜合以上實驗結果，我們發現年齡變項確實會影響政黨身分歸類作業以及特質評分作業的表現，且此影響可能會隨照片人物的政黨身分而有所不同。
People go beyond the inferences afforded by a person’s observable features to make inferences about personal traits or even social group membership. Rule and Ambady (2010) found that US college students can differentiate Republicans and Democrats by face. Here we extended Hu et al. (2016) study to further explore the effect of age on differentiating KMT (Kuomintang) and DPP (Democratic Progressive Party) candidates from their photos. The present study included three experiments. In Experiment 1: Perceived age rating task, we explored the perceived age between KMT and DPP and aimed to reduce the difference. In Experiment 2: Political membership categorization task, we divided stimulus from Experiment 1 into 4 blocks by their age range, in order to explore the effect of the perceived age of candidates’ photos, within the same party, on categorization task performance. In Experiment 3: Trait rating task, we explored the typical traits that may differentiate KMT from DPP, as well as the possible effects of participants’ age. The results showed that participants were able to differentiate KMT from DPP candidates, even when the marginal difference between KMT and DPP candidates’ perceived age was reduced. We also discovered that provincial is the core trait of differentiating the parties for both younger and older participants and facial maturity is another important differentiating trait for older participants. Our study successfully explored the effect of perceived age and the core trait on differentiating KMT and DPP candidates from their photos.
