diff --git a/backend/app/services/news_service.py b/backend/app/services/news_service.py
index 9bb1027..6dca097 100755
--- a/backend/app/services/news_service.py
+++ b/backend/app/services/news_service.py
@@ -488,13 +488,14 @@ class NewsPlatformService:
                 return valid
             logger.info(f"[广场新闻] {user.account} 今日文章校验后全部无效，转历史")
 
-        # ── Phase 2: 无今日新文章 → 从最新到旧顺序翻页（随机从某页开始）────
-        # 规则：从第1页（最新）开始依次往后，轮到哪页由当前小时决定，
-        #       保证不同时段覆盖不同页，模拟"从新到旧"逐步互动
+        # ── Phase 2: 无今日新文章 → 从最新(第1页)开始往旧顺序遍历 ────
+        # 规则：始终从第1页（最新）开始，按页顺序 1→2→3...往旧方向走
+        #       每小时推进一页，保证不同时段覆盖不同深度的文章
         logger.info(f"[广场新闻] {user.account} 无今日新文章，从最新向旧顺序翻页")
 
-        # 获取总页数
+        # 获取总页数（第1页同时也是数据源）
         total_pages = 1
+        page1_items = []
         try:
             async with httpx.AsyncClient(timeout=10) as _c:
                 _r = await _c.get(
@@ -506,41 +507,50 @@ class NewsPlatformService:
             if _d.get("code") in [0, 200]:
                 total_size = _d.get("data", {}).get("totalSize", 0)
                 total_pages = max(1, math.ceil(total_size / 50))
+                nd = _d.get("data", {})
+                page1_items = _filter(nd.get("data") or nd.get("list") or nd.get("records") or [])
         except Exception:
             pass
 
-        max_pages = min(total_pages, 10)   # 最多翻前10页
-        # 用小时数取模决定起始页，保证同一小时内不同用户分散在不同页
-        # 从最新(page=1)开始往旧的方向走
-        hour_slot = _dt.now().hour % max_pages
-        start_page = hour_slot + 1  # 1-based
+        max_pages = min(total_pages, 10)
+        # 用小时决定本轮目标页：0点→第1页(最新)，每小时推进一页，循环往复
+        hour_page = (_dt.now().hour % max_pages) + 1  # 1-based，范围 1~max_pages
 
         items = []
-        # 尝试从 start_page 开始，若该页为空则顺序往后再往前找
-        pages_to_try = list(range(start_page, max_pages + 1)) + list(range(1, start_page))
-        tried_page = start_page
-        for page in pages_to_try:
-            try:
-                async with httpx.AsyncClient(timeout=15) as c:
-                    r = await c.get(
-                        f"{biz}/business/member/square/list",
-                        headers=self._bearer(token),
-                        params=_build(page),
-                    )
-                if r.status_code == 200:
-                    d = r.json()
-                    if d.get("code") in [0, 200]:
-                        nd     = d.get("data", {})
-                        _items = nd.get("data") or nd.get("list") or nd.get("records") or []
-                        _items = _filter(_items)
-                        if _items:
-                            items = _items
-                            tried_page = page
-                            break
-            except Exception as e:
-                logger.error(f"[广场新闻-历史] {user.account} page={page}: {e}")
+        tried_page = 1
 
-        logger.info(f"[广场新闻] {user.account} 历史第{tried_page}页获取到 {len(items)} 条")
+        if hour_page == 1:
+            # 当前时段对应第1页，直接使用已获取的数据
+            items = page1_items
+            tried_page = 1
+        else:
+            # 先尝试当前时段对应的目标页
+            pages_to_try = [hour_page] + list(range(1, hour_page)) + list(range(hour_page + 1, max_pages + 1))
+            for page in pages_to_try:
+                if page == 1 and page1_items:
+                    items = page1_items
+                    tried_page = 1
+                    break
+                try:
+                    async with httpx.AsyncClient(timeout=15) as c:
+                        r = await c.get(
+                            f"{biz}/business/member/square/list",
+                            headers=self._bearer(token),
+                            params=_build(page),
+                        )
+                    if r.status_code == 200:
+                        d = r.json()
+                        if d.get("code") in [0, 200]:
+                            nd     = d.get("data", {})
+                            _items = _filter(nd.get("data") or nd.get("list") or nd.get("records") or [])
+                            if _items:
+                                items = _items
+                                tried_page = page
+                                break
+                except Exception as e:
+                    logger.error(f"[广场新闻-历史] {user.account} page={page}: {e}")
+
+        logger.info(f"[广场新闻] {user.account} 历史第{tried_page}/{max_pages}页获取到 {len(items)} 条（时段对应第{hour_page}页）")
 
         if not items:
             return []